diff --git a/Notebooks/02_data_wrangling.ipynb b/Notebooks/02_data_wrangling.ipynb index a52eb6c24..93632d82d 100644 --- a/Notebooks/02_data_wrangling.ipynb +++ b/Notebooks/02_data_wrangling.ipynb @@ -1,2813 +1,3222 @@ { - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# 2 Data wrangling" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.1 Contents\n", - "* [2 Data wrangling](#2_Data_wrangling)\n", - " * [2.1 Contents](#2.1_Contents)\n", - " * [2.2 Introduction](#2.2_Introduction)\n", - " * [2.2.1 Recap Of Data Science Problem](#2.2.1_Recap_Of_Data_Science_Problem)\n", - " * [2.2.2 Introduction To Notebook](#2.2.2_Introduction_To_Notebook)\n", - " * [2.3 Imports](#2.3_Imports)\n", - " * [2.4 Objectives](#2.4_Objectives)\n", - " * [2.5 Load The Ski Resort Data](#2.5_Load_The_Ski_Resort_Data)\n", - " * [2.6 Explore The Data](#2.6_Explore_The_Data)\n", - " * [2.6.1 Find Your Resort Of Interest](#2.6.1_Find_Your_Resort_Of_Interest)\n", - " * [2.6.2 Number Of Missing Values By Column](#2.6.2_Number_Of_Missing_Values_By_Column)\n", - " * [2.6.3 Categorical Features](#2.6.3_Categorical_Features)\n", - " * [2.6.3.1 Unique Resort Names](#2.6.3.1_Unique_Resort_Names)\n", - " * [2.6.3.2 Region And State](#2.6.3.2_Region_And_State)\n", - " * [2.6.3.3 Number of distinct regions and states](#2.6.3.3_Number_of_distinct_regions_and_states)\n", - " * [2.6.3.4 Distribution Of Resorts By Region And State](#2.6.3.4_Distribution_Of_Resorts_By_Region_And_State)\n", - " * [2.6.3.5 Distribution Of Ticket Price By State](#2.6.3.5_Distribution_Of_Ticket_Price_By_State)\n", - " * [2.6.3.5.1 Average weekend and weekday price by state](#2.6.3.5.1_Average_weekend_and_weekday_price_by_state)\n", - " * [2.6.3.5.2 Distribution of weekday and weekend price by state](#2.6.3.5.2_Distribution_of_weekday_and_weekend_price_by_state)\n", - " * [2.6.4 Numeric Features](#2.6.4_Numeric_Features)\n", - " * [2.6.4.1 Numeric data summary](#2.6.4.1_Numeric_data_summary)\n", - " * [2.6.4.2 Distributions Of Feature Values](#2.6.4.2_Distributions_Of_Feature_Values)\n", - " * [2.6.4.2.1 SkiableTerrain_ac](#2.6.4.2.1_SkiableTerrain_ac)\n", - " * [2.6.4.2.2 Snow Making_ac](#2.6.4.2.2_Snow_Making_ac)\n", - " * [2.6.4.2.3 fastEight](#2.6.4.2.3_fastEight)\n", - " * [2.6.4.2.4 fastSixes and Trams](#2.6.4.2.4_fastSixes_and_Trams)\n", - " * [2.7 Derive State-wide Summary Statistics For Our Market Segment](#2.7_Derive_State-wide_Summary_Statistics_For_Our_Market_Segment)\n", - " * [2.8 Drop Rows With No Price Data](#2.8_Drop_Rows_With_No_Price_Data)\n", - " * [2.9 Review distributions](#2.9_Review_distributions)\n", - " * [2.10 Population data](#2.10_Population_data)\n", - " * [2.11 Target Feature](#2.11_Target_Feature)\n", - " * [2.11.1 Number Of Missing Values By Row - Resort](#2.11.1_Number_Of_Missing_Values_By_Row_-_Resort)\n", - " * [2.12 Save data](#2.12_Save_data)\n", - " * [2.13 Summary](#2.13_Summary)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.2 Introduction" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This step focuses on collecting your data, organizing it, and making sure it's well defined. Paying attention to these tasks will pay off greatly later on. Some data cleaning can be done at this stage, but it's important not to be overzealous in your cleaning before you've explored the data to better understand it." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.2.1 Recap Of Data Science Problem" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The purpose of this data science project is to come up with a pricing model for ski resort tickets in our market segment. Big Mountain suspects it may not be maximizing its returns, relative to its position in the market. It also does not have a strong sense of what facilities matter most to visitors, particularly which ones they're most likely to pay more for. This project aims to build a predictive model for ticket price based on a number of facilities, or properties, boasted by resorts (*at the resorts).* \n", - "This model will be used to provide guidance for Big Mountain's pricing and future facility investment plans." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.2.2 Introduction To Notebook" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Notebooks grow organically as we explore our data. If you used paper notebooks, you could discover a mistake and cross out or revise some earlier work. Later work may give you a reason to revisit earlier work and explore it further. The great thing about Jupyter notebooks is that you can edit, add, and move cells around without needing to cross out figures or scrawl in the margin. However, this means you can lose track of your changes easily. If you worked in a regulated environment, the company may have a a policy of always dating entries and clearly crossing out any mistakes, with your initials and the date.\n", - "\n", - "**Best practice here is to commit your changes using a version control system such as Git.** Try to get into the habit of adding and committing your files to the Git repository you're working in after you save them. You're are working in a Git repository, right? If you make a significant change, save the notebook and commit it to Git. In fact, if you're about to make a significant change, it's a good idea to commit before as well. Then if the change is a mess, you've got the previous version to go back to.\n", - "\n", - "**Another best practice with notebooks is to try to keep them organized with helpful headings and comments.** Not only can a good structure, but associated headings help you keep track of what you've done and your current focus. Anyone reading your notebook will have a much easier time following the flow of work. Remember, that 'anyone' will most likely be you. Be kind to future you!\n", - "\n", - "In this notebook, note how we try to use well structured, helpful headings that frequently are self-explanatory, and we make a brief note after any results to highlight key takeaways. This is an immense help to anyone reading your notebook and it will greatly help you when you come to summarise your findings. **Top tip: jot down key findings in a final summary at the end of the notebook as they arise. You can tidy this up later.** This is a great way to ensure important results don't get lost in the middle of your notebooks." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "In this, and subsequent notebooks, there are coding tasks marked with `#Code task n#` with code to complete. The `___` will guide you to where you need to insert code." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.3 Imports" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Placing your imports all together at the start of your notebook means you only need to consult one place to check your notebook's dependencies. By all means import something 'in situ' later on when you're experimenting, but if the imported dependency ends up being kept, you should subsequently move the import statement here with the rest." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 1#\n", - "#Import pandas, matplotlib.pyplot, and seaborn in the correct lines below\n", - "import ___ as pd\n", - "import ___ as plt\n", - "import ___ as sns\n", - "import os\n", - "\n", - "from library.sb_utils import save_file\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.4 Objectives" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "There are some fundamental questions to resolve in this notebook before you move on.\n", - "\n", - "* Do you think you may have the data you need to tackle the desired question?\n", - " * Have you identified the required target value?\n", - " * Do you have potentially useful features?\n", - "* Do you have any fundamental issues with the data?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.5 Load The Ski Resort Data" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": {}, - "outputs": [], - "source": [ - "# the supplied CSV data file is the raw_data directory\n", - "ski_data = pd.read_csv('../raw_data/ski_resort_data.csv')" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Good first steps in auditing the data are the info method and displaying the first few records with head." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 2#\n", - "#Call the info method on ski_data to see a summary of the data\n", - "ski_data.___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "`AdultWeekday` is the price of an adult weekday ticket. `AdultWeekend` is the price of an adult weekend ticket. The other columns are potential features." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This immediately raises the question of what quantity will you want to model? You know you want to model the ticket price, but you realise there are two kinds of ticket price!" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "scrolled": true - }, - "outputs": [], - "source": [ - "#Code task 3#\n", - "#Call the head method on ski_data to print the first several rows of the data\n", - "ski_data.___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The output above suggests you've made a good start getting the ski resort data organized. You have plausible column headings. You can already see you have a missing value in the `fastEight` column" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.6 Explore The Data" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.6.1 Find Your Resort Of Interest" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Your resort of interest is called Big Mountain Resort. Check it's in the data:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 4#\n", - "#Filter the ski_data dataframe to display just the row for our resort with the name 'Big Mountain Resort'\n", - "#Hint: you will find that the transpose of the row will give a nicer output. DataFrame's do have a\n", - "#transpose method, but you can access this conveniently with the `T` property.\n", - "ski_data[ski_data.Name == ___].___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "It's good that your resort doesn't appear to have any missing values." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.6.2 Number Of Missing Values By Column" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Count the number of missing values in each column and sort them." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 5#\n", - "#Count (using `.sum()`) the number of missing values (`.isnull()`) in each column of \n", - "#ski_data as well as the percentages (using `.mean()` instead of `.sum()`).\n", - "#Order them (increasing or decreasing) using sort_values\n", - "#Call `pd.concat` to present these in a single table (DataFrame) with the helpful column names 'count' and '%'\n", - "missing = ___([ski_data.___.___, 100 * ski_data.___.___], axis=1)\n", - "missing.columns=[___, ___]\n", - "missing.___(by=___)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "`fastEight` has the most missing values, at just over 50%. Unfortunately, you see you're also missing quite a few of your desired target quantity, the ticket price, which is missing 15-16% of values. `AdultWeekday` is missing in a few more records than `AdultWeekend`. What overlap is there in these missing values? This is a question you'll want to investigate. You should also point out that `isnull()` is not the only indicator of missing data. Sometimes 'missingness' can be encoded, perhaps by a -1 or 999. Such values are typically chosen because they are \"obviously\" not genuine values. If you were capturing data on people's heights and weights but missing someone's height, you could certainly encode that as a 0 because no one has a height of zero (in any units). Yet such entries would not be revealed by `isnull()`. Here, you need a data dictionary and/or to spot such values as part of looking for outliers. Someone with a height of zero should definitely show up as an outlier!" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.6.3 Categorical Features" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "So far you've examined only the numeric features. Now you inspect categorical ones such as resort name and state. These are discrete entities. 'Alaska' is a name. Although names can be sorted alphabetically, it makes no sense to take the average of 'Alaska' and 'Arizona'. Similarly, 'Alaska' is before 'Arizona' only lexicographically; it is neither 'less than' nor 'greater than' 'Arizona'. As such, they tend to require different handling than strictly numeric quantities. Note, a feature _can_ be numeric but also categorical. For example, instead of giving the number of `fastEight` lifts, a feature might be `has_fastEights` and have the value 0 or 1 to denote absence or presence of such a lift. In such a case it would not make sense to take an average of this or perform other mathematical calculations on it. Although you digress a little to make a point, month numbers are also, strictly speaking, categorical features. Yes, when a month is represented by its number (1 for January, 2 for Februrary etc.) it provides a convenient way to graph trends over a year. And, arguably, there is some logical interpretation of the average of 1 and 3 (January and March) being 2 (February). However, clearly December of one years precedes January of the next and yet 12 as a number is not less than 1. The numeric quantities in the section above are truly numeric; they are the number of feet in the drop, or acres or years open or the amount of snowfall etc." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 6#\n", - "#Use ski_data's `select_dtypes` method to select columns of dtype 'object'\n", - "ski_data.___(___)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You saw earlier on that these three columns had no missing values. But are there any other issues with these columns? Sensible questions to ask here include:\n", - "\n", - "* Is `Name` (or at least a combination of Name/Region/State) unique?\n", - "* Is `Region` always the same as `state`?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.3.1 Unique Resort Names" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 7#\n", - "#Use pandas' Series method `value_counts` to find any duplicated resort names\n", - "ski_data['Name'].___.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You have a duplicated resort name: Crystal Mountain." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Q: 1** Is this resort duplicated if you take into account Region and/or state as well?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 8#\n", - "#Concatenate the string columns 'Name' and 'Region' and count the values again (as above)\n", - "(ski_data[___] + ', ' + ski_data[___]).___.head()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 9#\n", - "#Concatenate 'Name' and 'state' and count the values again (as above)\n", - "(ski_data[___] + ', ' + ski_data[___]).___.head()" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "**NB** because you know `value_counts()` sorts descending, you can use the `head()` method and know the rest of the counts must be 1." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**A: 1** Your answer here" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
NameRegionstatesummit_elevvertical_dropbase_elevtramsfastEightfastSixesfastQuads...LongestRun_miSkiableTerrain_acSnow Making_acdaysOpenLastYearyearsOpenaverageSnowfallAdultWeekdayAdultWeekendprojectedDaysOpenNightSkiing_ac
104Crystal MountainMichiganMichigan113237575700.001...0.3102.096.0120.063.0132.054.064.0135.056.0
295Crystal MountainWashingtonWashington7012310044001NaN22...2.52600.010.0NaN57.0486.099.099.0NaNNaN
\n", - "

2 rows × 27 columns

\n", - "
" + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "yW8MqdcgTNrU" + }, + "source": [ + "# 2 Data wrangling" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "CRMs1zpCTNrV" + }, + "source": [ + "## 2.1 Contents\n", + "* [2 Data wrangling](#2_Data_wrangling)\n", + " * [2.1 Contents](#2.1_Contents)\n", + " * [2.2 Introduction](#2.2_Introduction)\n", + " * [2.2.1 Recap Of Data Science Problem](#2.2.1_Recap_Of_Data_Science_Problem)\n", + " * [2.2.2 Introduction To Notebook](#2.2.2_Introduction_To_Notebook)\n", + " * [2.3 Imports](#2.3_Imports)\n", + " * [2.4 Objectives](#2.4_Objectives)\n", + " * [2.5 Load The Ski Resort Data](#2.5_Load_The_Ski_Resort_Data)\n", + " * [2.6 Explore The Data](#2.6_Explore_The_Data)\n", + " * [2.6.1 Find Your Resort Of Interest](#2.6.1_Find_Your_Resort_Of_Interest)\n", + " * [2.6.2 Number Of Missing Values By Column](#2.6.2_Number_Of_Missing_Values_By_Column)\n", + " * [2.6.3 Categorical Features](#2.6.3_Categorical_Features)\n", + " * [2.6.3.1 Unique Resort Names](#2.6.3.1_Unique_Resort_Names)\n", + " * [2.6.3.2 Region And State](#2.6.3.2_Region_And_State)\n", + " * [2.6.3.3 Number of distinct regions and states](#2.6.3.3_Number_of_distinct_regions_and_states)\n", + " * [2.6.3.4 Distribution Of Resorts By Region And State](#2.6.3.4_Distribution_Of_Resorts_By_Region_And_State)\n", + " * [2.6.3.5 Distribution Of Ticket Price By State](#2.6.3.5_Distribution_Of_Ticket_Price_By_State)\n", + " * [2.6.3.5.1 Average weekend and weekday price by state](#2.6.3.5.1_Average_weekend_and_weekday_price_by_state)\n", + " * [2.6.3.5.2 Distribution of weekday and weekend price by state](#2.6.3.5.2_Distribution_of_weekday_and_weekend_price_by_state)\n", + " * [2.6.4 Numeric Features](#2.6.4_Numeric_Features)\n", + " * [2.6.4.1 Numeric data summary](#2.6.4.1_Numeric_data_summary)\n", + " * [2.6.4.2 Distributions Of Feature Values](#2.6.4.2_Distributions_Of_Feature_Values)\n", + " * [2.6.4.2.1 SkiableTerrain_ac](#2.6.4.2.1_SkiableTerrain_ac)\n", + " * [2.6.4.2.2 Snow Making_ac](#2.6.4.2.2_Snow_Making_ac)\n", + " * [2.6.4.2.3 fastEight](#2.6.4.2.3_fastEight)\n", + " * [2.6.4.2.4 fastSixes and Trams](#2.6.4.2.4_fastSixes_and_Trams)\n", + " * [2.7 Derive State-wide Summary Statistics For Our Market Segment](#2.7_Derive_State-wide_Summary_Statistics_For_Our_Market_Segment)\n", + " * [2.8 Drop Rows With No Price Data](#2.8_Drop_Rows_With_No_Price_Data)\n", + " * [2.9 Review distributions](#2.9_Review_distributions)\n", + " * [2.10 Population data](#2.10_Population_data)\n", + " * [2.11 Target Feature](#2.11_Target_Feature)\n", + " * [2.11.1 Number Of Missing Values By Row - Resort](#2.11.1_Number_Of_Missing_Values_By_Row_-_Resort)\n", + " * [2.12 Save data](#2.12_Save_data)\n", + " * [2.13 Summary](#2.13_Summary)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "OvD_0DroTNrW" + }, + "source": [ + "## 2.2 Introduction" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "F76Xr_mSTNrW" + }, + "source": [ + "This step focuses on collecting your data, organizing it, and making sure it's well defined. Paying attention to these tasks will pay off greatly later on. Some data cleaning can be done at this stage, but it's important not to be overzealous in your cleaning before you've explored the data to better understand it." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VNzIumXfTNrW" + }, + "source": [ + "### 2.2.1 Recap Of Data Science Problem" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "TbVp-JNDTNrW" + }, + "source": [ + "The purpose of this data science project is to come up with a pricing model for ski resort tickets in our market segment. Big Mountain suspects it may not be maximizing its returns, relative to its position in the market. It also does not have a strong sense of what facilities matter most to visitors, particularly which ones they're most likely to pay more for. This project aims to build a predictive model for ticket price based on a number of facilities, or properties, boasted by resorts (*at the resorts).*\n", + "This model will be used to provide guidance for Big Mountain's pricing and future facility investment plans." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MFnOENRuTNrX" + }, + "source": [ + "### 2.2.2 Introduction To Notebook" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "XWXuohx2TNrX" + }, + "source": [ + "Notebooks grow organically as we explore our data. If you used paper notebooks, you could discover a mistake and cross out or revise some earlier work. Later work may give you a reason to revisit earlier work and explore it further. The great thing about Jupyter notebooks is that you can edit, add, and move cells around without needing to cross out figures or scrawl in the margin. However, this means you can lose track of your changes easily. If you worked in a regulated environment, the company may have a a policy of always dating entries and clearly crossing out any mistakes, with your initials and the date.\n", + "\n", + "**Best practice here is to commit your changes using a version control system such as Git.** Try to get into the habit of adding and committing your files to the Git repository you're working in after you save them. You're are working in a Git repository, right? If you make a significant change, save the notebook and commit it to Git. In fact, if you're about to make a significant change, it's a good idea to commit before as well. Then if the change is a mess, you've got the previous version to go back to.\n", + "\n", + "**Another best practice with notebooks is to try to keep them organized with helpful headings and comments.** Not only can a good structure, but associated headings help you keep track of what you've done and your current focus. Anyone reading your notebook will have a much easier time following the flow of work. Remember, that 'anyone' will most likely be you. Be kind to future you!\n", + "\n", + "In this notebook, note how we try to use well structured, helpful headings that frequently are self-explanatory, and we make a brief note after any results to highlight key takeaways. This is an immense help to anyone reading your notebook and it will greatly help you when you come to summarise your findings. **Top tip: jot down key findings in a final summary at the end of the notebook as they arise. You can tidy this up later.** This is a great way to ensure important results don't get lost in the middle of your notebooks." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "t0HANEb0TNrX" + }, + "source": [ + "In this, and subsequent notebooks, there are coding tasks marked with `#Code task n#` with code to complete. The `___` will guide you to where you need to insert code." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zwBgG09yTNrX" + }, + "source": [ + "## 2.3 Imports" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6Fq2Zn5QTNrX" + }, + "source": [ + "Placing your imports all together at the start of your notebook means you only need to consult one place to check your notebook's dependencies. By all means import something 'in situ' later on when you're experimenting, but if the imported dependency ends up being kept, you should subsequently move the import statement here with the rest." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "wmP-dUztTNrX" + }, + "outputs": [], + "source": [ + "#Code task 1#\n", + "#Import pandas, matplotlib.pyplot, and seaborn in the correct lines below\n", + "import ___ as pd\n", + "import ___ as plt\n", + "import ___ as sns\n", + "import os\n", + "\n", + "from library.sb_utils import save_file\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UQGE01sCTNrY" + }, + "source": [ + "## 2.4 Objectives" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "my0jzNgITNrY" + }, + "source": [ + "There are some fundamental questions to resolve in this notebook before you move on.\n", + "\n", + "* Do you think you may have the data you need to tackle the desired question?\n", + " * Have you identified the required target value?\n", + " * Do you have potentially useful features?\n", + "* Do you have any fundamental issues with the data?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "AJ_yDDTOTNrY" + }, + "source": [ + "## 2.5 Load The Ski Resort Data" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "xisHZeYQTNrY" + }, + "outputs": [], + "source": [ + "# the supplied CSV data file is the raw_data directory\n", + "ski_data = pd.read_csv('../raw_data/ski_resort_data.csv')" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qsrYRwzgTNrZ" + }, + "source": [ + "Good first steps in auditing the data are the info method and displaying the first few records with head." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "FKJF5TsPTNrZ" + }, + "outputs": [], + "source": [ + "#Code task 2#\n", + "#Call the info method on ski_data to see a summary of the data\n", + "ski_data.___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UdmiwzsfTNrZ" + }, + "source": [ + "`AdultWeekday` is the price of an adult weekday ticket. `AdultWeekend` is the price of an adult weekend ticket. The other columns are potential features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1qYz_90dTNrZ" + }, + "source": [ + "This immediately raises the question of what quantity will you want to model? You know you want to model the ticket price, but you realise there are two kinds of ticket price!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true, + "id": "dIV0jMHPTNrZ" + }, + "outputs": [], + "source": [ + "#Code task 3#\n", + "#Call the head method on ski_data to print the first several rows of the data\n", + "ski_data.___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1jVulon8TNrZ" + }, + "source": [ + "The output above suggests you've made a good start getting the ski resort data organized. You have plausible column headings. You can already see you have a missing value in the `fastEight` column" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "XvlkYub_TNrZ" + }, + "source": [ + "## 2.6 Explore The Data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "YY9MZE8STNrZ" + }, + "source": [ + "### 2.6.1 Find Your Resort Of Interest" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9Oa0yq6wTNra" + }, + "source": [ + "Your resort of interest is called Big Mountain Resort. Check it's in the data:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "X3Mxy8QYTNra" + }, + "outputs": [], + "source": [ + "#Code task 4#\n", + "#Filter the ski_data dataframe to display just the row for our resort with the name 'Big Mountain Resort'\n", + "#Hint: you will find that the transpose of the row will give a nicer output. DataFrame's do have a\n", + "#transpose method, but you can access this conveniently with the `T` property.\n", + "ski_data[ski_data.Name == ___].___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "W0BMZnjnTNra" + }, + "source": [ + "It's good that your resort doesn't appear to have any missing values." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "lclEAmviTNra" + }, + "source": [ + "### 2.6.2 Number Of Missing Values By Column" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Z1RYZRtlTNra" + }, + "source": [ + "Count the number of missing values in each column and sort them." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "V49AlHZoTNra" + }, + "outputs": [], + "source": [ + "#Code task 5#\n", + "#Count (using `.sum()`) the number of missing values (`.isnull()`) in each column of\n", + "#ski_data as well as the percentages (using `.mean()` instead of `.sum()`).\n", + "#Order them (increasing or decreasing) using sort_values\n", + "#Call `pd.concat` to present these in a single table (DataFrame) with the helpful column names 'count' and '%'\n", + "missing = ___([ski_data.___.___, 100 * ski_data.___.___], axis=1)\n", + "missing.columns=[___, ___]\n", + "missing.___(by=___)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Id9MvWO4TNra" + }, + "source": [ + "`fastEight` has the most missing values, at just over 50%. Unfortunately, you see you're also missing quite a few of your desired target quantity, the ticket price, which is missing 15-16% of values. `AdultWeekday` is missing in a few more records than `AdultWeekend`. What overlap is there in these missing values? This is a question you'll want to investigate. You should also point out that `isnull()` is not the only indicator of missing data. Sometimes 'missingness' can be encoded, perhaps by a -1 or 999. Such values are typically chosen because they are \"obviously\" not genuine values. If you were capturing data on people's heights and weights but missing someone's height, you could certainly encode that as a 0 because no one has a height of zero (in any units). Yet such entries would not be revealed by `isnull()`. Here, you need a data dictionary and/or to spot such values as part of looking for outliers. Someone with a height of zero should definitely show up as an outlier!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qzmZ5vf3TNra" + }, + "source": [ + "### 2.6.3 Categorical Features" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "sWxiJWPkTNrb" + }, + "source": [ + "So far you've examined only the numeric features. Now you inspect categorical ones such as resort name and state. These are discrete entities. 'Alaska' is a name. Although names can be sorted alphabetically, it makes no sense to take the average of 'Alaska' and 'Arizona'. Similarly, 'Alaska' is before 'Arizona' only lexicographically; it is neither 'less than' nor 'greater than' 'Arizona'. As such, they tend to require different handling than strictly numeric quantities. Note, a feature _can_ be numeric but also categorical. For example, instead of giving the number of `fastEight` lifts, a feature might be `has_fastEights` and have the value 0 or 1 to denote absence or presence of such a lift. In such a case it would not make sense to take an average of this or perform other mathematical calculations on it. Although you digress a little to make a point, month numbers are also, strictly speaking, categorical features. Yes, when a month is represented by its number (1 for January, 2 for Februrary etc.) it provides a convenient way to graph trends over a year. And, arguably, there is some logical interpretation of the average of 1 and 3 (January and March) being 2 (February). However, clearly December of one years precedes January of the next and yet 12 as a number is not less than 1. The numeric quantities in the section above are truly numeric; they are the number of feet in the drop, or acres or years open or the amount of snowfall etc." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "7H0myqwCTNrb" + }, + "outputs": [], + "source": [ + "#Code task 6#\n", + "#Use ski_data's `select_dtypes` method to select columns of dtype 'object'\n", + "ski_data.___(___)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "VCXm808NTNrb" + }, + "source": [ + "You saw earlier on that these three columns had no missing values. But are there any other issues with these columns? Sensible questions to ask here include:\n", + "\n", + "* Is `Name` (or at least a combination of Name/Region/State) unique?\n", + "* Is `Region` always the same as `state`?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rMouUH1aTNrb" + }, + "source": [ + "#### 2.6.3.1 Unique Resort Names" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "IrZxYtdjTNrb" + }, + "outputs": [], + "source": [ + "#Code task 7#\n", + "#Use pandas' Series method `value_counts` to find any duplicated resort names\n", + "ski_data['Name'].___.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZiobtpQ7TNrb" + }, + "source": [ + "You have a duplicated resort name: Crystal Mountain." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QFIRzh1UTNrb" + }, + "source": [ + "**Q: 1** Is this resort duplicated if you take into account Region and/or state as well?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "IdcW8a8dTNrc" + }, + "outputs": [], + "source": [ + "#Code task 8#\n", + "#Concatenate the string columns 'Name' and 'Region' and count the values again (as above)\n", + "(ski_data[___] + ', ' + ski_data[___]).___.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "x8yHtL7rTNrc" + }, + "outputs": [], + "source": [ + "#Code task 9#\n", + "#Concatenate 'Name' and 'state' and count the values again (as above)\n", + "(ski_data[___] + ', ' + ski_data[___]).___.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "2Aq40cySTNrc" + }, + "outputs": [], + "source": [ + "**NB** because you know `value_counts()` sorts descending, you can use the `head()` method and know the rest of the counts must be 1." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "CmcugecHTNrc" + }, + "source": [ + "**A: 1** Your answer here" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "78boDJS2TNrc", + "outputId": "57b4c86f-354d-4afc-ae4b-0470031f44f3" + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
NameRegionstatesummit_elevvertical_dropbase_elevtramsfastEightfastSixesfastQuads...LongestRun_miSkiableTerrain_acSnow Making_acdaysOpenLastYearyearsOpenaverageSnowfallAdultWeekdayAdultWeekendprojectedDaysOpenNightSkiing_ac
104Crystal MountainMichiganMichigan113237575700.001...0.3102.096.0120.063.0132.054.064.0135.056.0
295Crystal MountainWashingtonWashington7012310044001NaN22...2.52600.010.0NaN57.0486.099.099.0NaNNaN
\n", + "

2 rows × 27 columns

\n", + "
" + ], + "text/plain": [ + " Name Region state summit_elev vertical_drop \\\n", + "104 Crystal Mountain Michigan Michigan 1132 375 \n", + "295 Crystal Mountain Washington Washington 7012 3100 \n", + "\n", + " base_elev trams fastEight fastSixes fastQuads ... LongestRun_mi \\\n", + "104 757 0 0.0 0 1 ... 0.3 \n", + "295 4400 1 NaN 2 2 ... 2.5 \n", + "\n", + " SkiableTerrain_ac Snow Making_ac daysOpenLastYear yearsOpen \\\n", + "104 102.0 96.0 120.0 63.0 \n", + "295 2600.0 10.0 NaN 57.0 \n", + "\n", + " averageSnowfall AdultWeekday AdultWeekend projectedDaysOpen \\\n", + "104 132.0 54.0 64.0 135.0 \n", + "295 486.0 99.0 99.0 NaN \n", + "\n", + " NightSkiing_ac \n", + "104 56.0 \n", + "295 NaN \n", + "\n", + "[2 rows x 27 columns]" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data[ski_data['Name'] == 'Crystal Mountain']" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "nCqBbQsHTNrd" + }, + "source": [ + "So there are two Crystal Mountain resorts, but they are clearly two different resorts in two different states. This is a powerful signal that you have unique records on each row." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "eHUrzGTmTNrd" + }, + "source": [ + "#### 2.6.3.2 Region And State" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rpzdEtnhTNrd" + }, + "source": [ + "What's the relationship between region and state?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "q7b1Np25TNrd" + }, + "source": [ + "You know they are the same in many cases (e.g. both the Region and the state are given as 'Michigan'). In how many cases do they differ?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "9hUIrnQuTNre" + }, + "outputs": [], + "source": [ + "#Code task 10#\n", + "#Calculate the number of times Region does not equal state\n", + "(ski_data.Region ___ ski_data.state).___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IrpzJI_5TNre" + }, + "source": [ + "You know what a state is. What is a region? You can tabulate the distinct values along with their respective frequencies using `value_counts()`." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "N0dlaCC5TNre", + "outputId": "5f6d91af-abb4-4d64-c9aa-1df358fe0e12" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "New York 33\n", + "Michigan 29\n", + "Sierra Nevada 22\n", + "Colorado 22\n", + "Pennsylvania 19\n", + "Wisconsin 16\n", + "New Hampshire 16\n", + "Vermont 15\n", + "Minnesota 14\n", + "Montana 12\n", + "Idaho 12\n", + "Massachusetts 11\n", + "Washington 10\n", + "Maine 9\n", + "New Mexico 9\n", + "Wyoming 8\n", + "Utah 7\n", + "Oregon 6\n", + "Salt Lake City 6\n", + "North Carolina 6\n", + "Connecticut 5\n", + "Ohio 5\n", + "West Virginia 4\n", + "Virginia 4\n", + "Mt. Hood 4\n", + "Illinois 4\n", + "Alaska 3\n", + "Iowa 3\n", + "Missouri 2\n", + "Arizona 2\n", + "Indiana 2\n", + "South Dakota 2\n", + "New Jersey 2\n", + "Nevada 2\n", + "Rhode Island 1\n", + "Maryland 1\n", + "Tennessee 1\n", + "Northern California 1\n", + "Name: Region, dtype: int64" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data['Region'].value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6MMZVyQQTNre" + }, + "source": [ + "A casual inspection by eye reveals some non-state names such as Sierra Nevada, Salt Lake City, and Northern California. Tabulate the differences between Region and state. On a note regarding scaling to larger data sets, you might wonder how you could spot such cases when presented with millions of rows. This is an interesting point. Imagine you have access to a database with a Region and state column in a table and there are millions of rows. You wouldn't eyeball all the rows looking for differences! Bear in mind that our first interest lies in establishing the answer to the question \"Are they always the same?\" One approach might be to ask the database to return records where they differ, but limit the output to 10 rows. If there were differences, you'd only get up to 10 results, and so you wouldn't know whether you'd located all differences, but you'd know that there were 'a nonzero number' of differences. If you got an empty result set back, then you would know that the two columns always had the same value. At the risk of digressing, some values in one column only might be NULL (missing) and different databases treat NULL differently, so be aware that on many an occasion a seamingly 'simple' question gets very interesting to answer very quickly!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "WXS6N7zYTNre" + }, + "outputs": [], + "source": [ + "#Code task 11#\n", + "#Filter the ski_data dataframe for rows where 'Region' and 'state' are different,\n", + "#group that by 'state' and perform `value_counts` on the 'Region'\n", + "(ski_data[ski_data.___ ___ ski_data.___]\n", + " .groupby(___)[___]\n", + " .value_counts())" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9tNJQGgxTNre" + }, + "source": [ + "The vast majority of the differences are in California, with most Regions being called Sierra Nevada and just one referred to as Northern California." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "wi6pbzZVTNre" + }, + "source": [ + "#### 2.6.3.3 Number of distinct regions and states" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ySPFicDkTNrf" + }, + "outputs": [], + "source": [ + "#Code task 12#\n", + "#Select the 'Region' and 'state' columns from ski_data and use the `nunique` method to calculate\n", + "#the number of unique values in each\n", + "ski_data[[___, ___]].___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "OZldLtq7TNrf" + }, + "source": [ + "Because a few states are split across multiple named regions, there are slightly more unique regions than states." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qLB88KkkTNrf" + }, + "source": [ + "#### 2.6.3.4 Distribution Of Resorts By Region And State" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "LkOpFlVvTNrf" + }, + "source": [ + "If this is your first time using [matplotlib](https://matplotlib.org/3.2.2/index.html)'s [subplots](https://matplotlib.org/3.2.2/api/_as_gen/matplotlib.pyplot.subplots.html), you may find the online documentation useful." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "9IqGsH_QTNrf" + }, + "outputs": [], + "source": [ + "#Code task 13#\n", + "#Create two subplots on 1 row and 2 columns with a figsize of (12, 8)\n", + "fig, ax = plt.subplots(___, ___, figsize=(___))\n", + "#Specify a horizontal barplot ('barh') as kind of plot (kind=)\n", + "ski_data.Region.value_counts().plot(kind=___, ax=ax[0])\n", + "#Give the plot a helpful title of 'Region'\n", + "ax[0].set_title(___)\n", + "#Label the xaxis 'Count'\n", + "ax[0].set_xlabel(___)\n", + "#Specify a horizontal barplot ('barh') as kind of plot (kind=)\n", + "ski_data.state.value_counts().plot(kind=___, ax=ax[1])\n", + "#Give the plot a helpful title of 'state'\n", + "ax[1].set_title(___)\n", + "#Label the xaxis 'Count'\n", + "ax[1].set_xlabel(___)\n", + "#Give the subplots a little \"breathing room\" with a wspace of 0.5\n", + "plt.subplots_adjust(wspace=___);\n", + "#You're encouraged to explore a few different figure sizes, orientations, and spacing here\n", + "# as the importance of easy-to-read and informative figures is frequently understated\n", + "# and you will find the ability to tweak figures invaluable later on" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dpijziQsTNrf" + }, + "source": [ + "How's your geography? Looking at the distribution of States, you see New York accounting for the majority of resorts. Our target resort is in Montana, which comes in at 13th place. You should think carefully about how, or whether, you use this information. Does New York command a premium because of its proximity to population? Even if a resort's State were a useful predictor of ticket price, your main interest lies in Montana. Would you want a model that is skewed for accuracy by New York? Should you just filter for Montana and create a Montana-specific model? This would slash your available data volume. Your problem task includes the contextual insight that the data are for resorts all belonging to the same market share. This suggests one might expect prices to be similar amongst them. You can look into this. A boxplot grouped by State is an ideal way to quickly compare prices. Another side note worth bringing up here is that, in reality, the best approach here definitely would include consulting with the client or other domain expert. They might know of good reasons for treating states equivalently or differently. The data scientist is rarely the final arbiter of such a decision. But here, you'll see if we can find any supporting evidence for treating states the same or differently." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "HFjXIvmkTNrg" + }, + "source": [ + "#### 2.6.3.5 Distribution Of Ticket Price By State" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5shVIdb8TNrg" + }, + "source": [ + "Our primary focus is our Big Mountain resort, in Montana. Does the state give you any clues to help decide what your primary target response feature should be (weekend or weekday ticket prices)?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4mxWiGzkTNrg" + }, + "source": [ + "##### 2.6.3.5.1 Average weekend and weekday price by state" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "71TDNysNTNrg" + }, + "outputs": [], + "source": [ + "#Code task 14#\n", + "# Calculate average weekday and weekend price by state and sort by the average of the two\n", + "# Hint: use the pattern dataframe.groupby()[].mean()\n", + "state_price_means = ski_data.___(___)[[___, ___]].mean()\n", + "state_price_means.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "v7e2O_eDTNrg", + "outputId": "bd888d63-92b6-4dfd-8ae0-8e8d0d77c626" + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# The next bit simply reorders the index by increasing average of weekday and weekend prices\n", + "# Compare the index order you get from\n", + "# state_price_means.index\n", + "# with\n", + "# state_price_means.mean(axis=1).sort_values(ascending=False).index\n", + "# See how this expression simply sits within the reindex()\n", + "(state_price_means.reindex(index=state_price_means.mean(axis=1)\n", + " .sort_values(ascending=False)\n", + " .index)\n", + " .plot(kind='barh', figsize=(10, 10), title='Average ticket price by State'))\n", + "plt.xlabel('Price ($)');" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "9whuYMlDTNrg" + }, + "outputs": [], + "source": [ + "The figure above represents a dataframe with two columns, one for the average prices of each kind of ticket. This tells you how the average ticket price varies from state to state. But can you get more insight into the difference in the distributions between states?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "x56e56BMTNrh" + }, + "source": [ + "##### 2.6.3.5.2 Distribution of weekday and weekend price by state" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hOkBuilFTNrh" + }, + "source": [ + "Next, you can transform the data into a single column for price with a new categorical column that represents the ticket type." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "jBFhADcGTNrh" + }, + "outputs": [], + "source": [ + "#Code task 15#\n", + "#Use the pd.melt function, pass in the ski_data columns 'state', 'AdultWeekday', and 'Adultweekend' only,\n", + "#specify 'state' for `id_vars`\n", + "#gather the ticket prices from the 'Adultweekday' and 'AdultWeekend' columns using the `value_vars` argument,\n", + "#call the resultant price column 'Price' via the `value_name` argument,\n", + "#name the weekday/weekend indicator column 'Ticket' via the `var_name` argument\n", + "ticket_prices = pd.melt(ski_data[[___, ___, ___]],\n", + " id_vars=___,\n", + " var_name=___,\n", + " value_vars=[___, ___],\n", + " value_name=___)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "GNS1wSepTNrh", + "outputId": "95e16415-2baf-4823-daac-ea323a26523f" + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
stateTicketPrice
0AlaskaAdultWeekday65.0
1AlaskaAdultWeekday47.0
2AlaskaAdultWeekday30.0
3ArizonaAdultWeekday89.0
4ArizonaAdultWeekday74.0
\n", + "
" + ], + "text/plain": [ + " state Ticket Price\n", + "0 Alaska AdultWeekday 65.0\n", + "1 Alaska AdultWeekday 47.0\n", + "2 Alaska AdultWeekday 30.0\n", + "3 Arizona AdultWeekday 89.0\n", + "4 Arizona AdultWeekday 74.0" + ] + }, + "execution_count": 20, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ticket_prices.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qgto7eHrTNrh" + }, + "source": [ + "This is now in a format we can pass to [seaborn](https://seaborn.pydata.org/)'s [boxplot](https://seaborn.pydata.org/generated/seaborn.boxplot.html) function to create boxplots of the ticket price distributions for each ticket type for each state." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mQeOgyu1TNrh" + }, + "outputs": [], + "source": [ + "#Code task 16#\n", + "#Create a seaborn boxplot of the ticket price dataframe we created above,\n", + "#with 'state' on the x-axis, 'Price' as the y-value, and a hue that indicates 'Ticket'\n", + "#This will use boxplot's x, y, hue, and data arguments.\n", + "plt.subplots(figsize=(12, 8))\n", + "sns.boxplot(x=___, y=___, hue=___, data=ticket_prices)\n", + "plt.xticks(rotation='vertical')\n", + "plt.ylabel('Price ($)')\n", + "plt.xlabel('State');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1QZWGbswTNri" + }, + "source": [ + "Aside from some relatively expensive ticket prices in California, Colorado, and Utah, most prices appear to lie in a broad band from around 25 to over 100 dollars. Some States show more variability than others. Montana and South Dakota, for example, both show fairly small variability as well as matching weekend and weekday ticket prices. Nevada and Utah, on the other hand, show the most range in prices. Some States, notably North Carolina and Virginia, have weekend prices far higher than weekday prices. You could be inspired from this exploration to consider a few potential groupings of resorts, those with low spread, those with lower averages, and those that charge a premium for weekend tickets. However, you're told that you are taking all resorts to be part of the same market share, you could argue against further segment the resorts. Nevertheless, ways to consider using the State information in your modelling include:\n", + "\n", + "* disregard State completely\n", + "* retain all State information\n", + "* retain State in the form of Montana vs not Montana, as our target resort is in Montana\n", + "\n", + "You've also noted another effect above: some States show a marked difference between weekday and weekend ticket prices. It may make sense to allow a model to take into account not just State but also weekend vs weekday." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3jwdl3TJTNri" + }, + "source": [ + "Thus we currently have two main questions you want to resolve:\n", + "\n", + "* What do you do about the two types of ticket price?\n", + "* What do you do about the state information?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PtkcCq6sTNri" + }, + "source": [ + "### 2.6.4 Numeric Features" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "WfTJNU5NTNri" + }, + "outputs": [], + "source": [ + "Having decided to reserve judgement on how exactly you utilize the State, turn your attention to cleaning the numeric features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "yy7EAty8TNri" + }, + "source": [ + "#### 2.6.4.1 Numeric data summary" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "yry0HlbYTNri" + }, + "outputs": [], + "source": [ + "#Code task 17#\n", + "#Call ski_data's `describe` method for a statistical summary of the numerical columns\n", + "#Hint: there are fewer summary stat columns than features, so displaying the transpose\n", + "#will be useful again\n", + "ski_data.___.___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "nhQ5Aa0qTNri" + }, + "source": [ + "Recall you're missing the ticket prices for some 16% of resorts. This is a fundamental problem that means you simply lack the required data for those resorts and will have to drop those records. But you may have a weekend price and not a weekday price, or vice versa. You want to keep any price you have." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Cgv8jEibTNrj", + "outputId": "54dbe828-1345-4564-c6cf-17f04b98c5cd" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "0 82.424242\n", + "2 14.242424\n", + "1 3.333333\n", + "dtype: float64" + ] + }, + "execution_count": 23, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "missing_price = ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum(axis=1)\n", + "missing_price.value_counts()/len(missing_price) * 100" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "oDWn1WBGTNrj" + }, + "source": [ + "Just over 82% of resorts have no missing ticket price, 3% are missing one value, and 14% are missing both. You will definitely want to drop the records for which you have no price information, however you will not do so just yet. There may still be useful information about the distributions of other features in that 14% of the data." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "TlwdBkSPTNrj" + }, + "source": [ + "#### 2.6.4.2 Distributions Of Feature Values" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jNIPE2oLTNrj" + }, + "source": [ + "Note that, although we are still in the 'data wrangling and cleaning' phase rather than exploratory data analysis, looking at distributions of features is immensely useful in getting a feel for whether the values look sensible and whether there are any obvious outliers to investigate. Some exploratory data analysis belongs here, and data wrangling will inevitably occur later on. It's more a matter of emphasis. Here, we're interesting in focusing on whether distributions look plausible or wrong. Later on, we're more interested in relationships and patterns." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "rZwEb5ZfTNrj" + }, + "outputs": [], + "source": [ + "#Code task 18#\n", + "#Call ski_data's `hist` method to plot histograms of each of the numeric features\n", + "#Try passing it an argument figsize=(15,10)\n", + "#Try calling plt.subplots_adjust() with an argument hspace=0.5 to adjust the spacing\n", + "#It's important you create legible and easy-to-read plots\n", + "ski_data.___(___)\n", + "#plt.subplots_adjust(hspace=___);\n", + "#Hint: notice how the terminating ';' \"swallows\" some messy output and leads to a tidier notebook" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "swCNX5c1TNrk" + }, + "source": [ + "What features do we have possible cause for concern about and why?\n", + "\n", + "* SkiableTerrain_ac because values are clustered down the low end,\n", + "* Snow Making_ac for the same reason,\n", + "* fastEight because all but one value is 0 so it has very little variance, and half the values are missing,\n", + "* fastSixes raises an amber flag; it has more variability, but still mostly 0,\n", + "* trams also may get an amber flag for the same reason,\n", + "* yearsOpen because most values are low but it has a maximum of 2019, which strongly suggests someone recorded calendar year rather than number of years." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PiYnWZ_ZTNrk" + }, + "source": [ + "##### 2.6.4.2.1 SkiableTerrain_ac" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mQ24ahTCTNrk" + }, + "outputs": [], + "source": [ + "#Code task 19#\n", + "#Filter the 'SkiableTerrain_ac' column to print the values greater than 10000\n", + "ski_data.___[ski_data.___ > ___]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "haQPnHRVTNrk" + }, + "source": [ + "**Q: 2** One resort has an incredibly large skiable terrain area! Which is it?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "AdoTV36ZTNrk" + }, + "outputs": [], + "source": [ + "#Code task 20#\n", + "#Now you know there's only one, print the whole row to investigate all values, including seeing the resort name\n", + "#Hint: don't forget the transpose will be helpful here\n", + "ski_data[ski_data.___ > ___].___" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DVhhh6U0TNrk" + }, + "source": [ + "**A: 2** Your answer here" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "g3Eo7Il5TNrk" + }, + "source": [ + "But what can you do when you have one record that seems highly suspicious?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hbLhuhbiTNrl" + }, + "source": [ + "You can see if your data are correct. Search for \"silverton mountain skiable area\". If you do this, you get some [useful information](https://www.google.com/search?q=silverton+mountain+skiable+area)." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "debj8lCRTNrl" + }, + "source": [ + "![Silverton Mountain information](https://github.com/JLindsey96/DataScienceGuidedCapstone/blob/master/Notebooks/images/silverton_mountain_info.png?raw=1)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mdDMB5_MTNrl" + }, + "source": [ + "You can spot check data. You see your top and base elevation values agree, but the skiable area is very different. Your suspect value is 26819, but the value you've just looked up is 1819. The last three digits agree. This sort of error could have occured in transmission or some editing or transcription stage. You could plausibly replace the suspect value with the one you've just obtained. Another cautionary note to make here is that although you're doing this in order to progress with your analysis, this is most definitely an issue that should have been raised and fed back to the client or data originator as a query. You should view this \"data correction\" step as a means to continue (documenting it carefully as you do in this notebook) rather than an ultimate decision as to what is correct." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3Vy4aBRtTNrl" + }, + "outputs": [], + "source": [ + "#Code task 21#\n", + "#Use the .loc accessor to print the 'SkiableTerrain_ac' value only for this resort\n", + "ski_data.___[39, 'SkiableTerrain_ac']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gGbLsiHzTNrl" + }, + "outputs": [], + "source": [ + "#Code task 22#\n", + "#Use the .loc accessor again to modify this value with the correct value of 1819\n", + "ski_data.___[39, 'SkiableTerrain_ac'] = ___" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mLECCwCrTNrl" + }, + "outputs": [], + "source": [ + "#Code task 23#\n", + "#Use the .loc accessor a final time to verify that the value has been modified\n", + "ski_data.___[39, 'SkiableTerrain_ac']" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mguQYF9STNrl" + }, + "source": [ + "**NB whilst you may become suspicious about your data quality, and you know you have missing values, you will not here dive down the rabbit hole of checking all values or web scraping to replace missing values.**" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1ra2paEeTNrl" + }, + "source": [ + "What does the distribution of skiable area look like now?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "pW-oHhT1TNrm", + "outputId": "dc4dcb34-9668-468f-8cd0-583df4d049e9" + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "ski_data.SkiableTerrain_ac.hist(bins=30)\n", + "plt.xlabel('SkiableTerrain_ac')\n", + "plt.ylabel('Count')\n", + "plt.title('Distribution of skiable area (acres) after replacing erroneous value');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zCSCIMbDTNrm" + }, + "source": [ + "You now see a rather long tailed distribution. You may wonder about the now most extreme value that is above 8000, but similarly you may also wonder about the value around 7000. If you wanted to spend more time manually checking values you could, but leave this for now. The above distribution is plausible." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "wK6YiCepTNrm" + }, + "source": [ + "##### 2.6.4.2.2 Snow Making_ac" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "I-bALhBJTNrm", + "outputId": "b8985868-8a34-4210-d771-14b9820a3b67" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "11 3379.0\n", + "18 1500.0\n", + "Name: Snow Making_ac, dtype: float64" + ] + }, + "execution_count": 31, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data['Snow Making_ac'][ski_data['Snow Making_ac'] > 1000]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Dv32rpWGTNrm", + "outputId": "d74d0faa-ec61-453d-bbbc-a8eb99ad8aeb" + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
11
NameHeavenly Mountain Resort
RegionSierra Nevada
stateCalifornia
summit_elev10067
vertical_drop3500
base_elev7170
trams2
fastEight0
fastSixes2
fastQuads7
quad1
triple5
double3
surface8
total_chairs28
Runs97
TerrainParks3
LongestRun_mi5.5
SkiableTerrain_ac4800
Snow Making_ac3379
daysOpenLastYear155
yearsOpen64
averageSnowfall360
AdultWeekdayNaN
AdultWeekendNaN
projectedDaysOpen157
NightSkiing_acNaN
\n", + "
" + ], + "text/plain": [ + " 11\n", + "Name Heavenly Mountain Resort\n", + "Region Sierra Nevada\n", + "state California\n", + "summit_elev 10067\n", + "vertical_drop 3500\n", + "base_elev 7170\n", + "trams 2\n", + "fastEight 0\n", + "fastSixes 2\n", + "fastQuads 7\n", + "quad 1\n", + "triple 5\n", + "double 3\n", + "surface 8\n", + "total_chairs 28\n", + "Runs 97\n", + "TerrainParks 3\n", + "LongestRun_mi 5.5\n", + "SkiableTerrain_ac 4800\n", + "Snow Making_ac 3379\n", + "daysOpenLastYear 155\n", + "yearsOpen 64\n", + "averageSnowfall 360\n", + "AdultWeekday NaN\n", + "AdultWeekend NaN\n", + "projectedDaysOpen 157\n", + "NightSkiing_ac NaN" + ] + }, + "execution_count": 32, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data[ski_data['Snow Making_ac'] > 3000].T" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "R5xk7E7HTNrn" + }, + "source": [ + "You can adopt a similar approach as for the suspect skiable area value and do some spot checking. To save time, here is a link to the website for [Heavenly Mountain Resort](https://www.skiheavenly.com/the-mountain/about-the-mountain/mountain-info.aspx). From this you can glean that you have values for skiable terrain that agree. Furthermore, you can read that snowmaking covers 60% of the trails." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "XRWd1-btTNrn" + }, + "source": [ + "What, then, is your rough guess for the area covered by snowmaking?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "lrMU2cq4TNrn", + "outputId": "f7bb07b0-1c60-405e-acb0-151363980fe6" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "2880.0" + ] + }, + "execution_count": 33, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + ".6 * 4800" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xelkMGxrTNrn" + }, + "source": [ + "This is less than the value of 3379 in your data so you may have a judgement call to make. However, notice something else. You have no ticket pricing information at all for this resort. Any further effort spent worrying about values for this resort will be wasted. You'll simply be dropping the entire row!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BtnIe-skTNrn" + }, + "source": [ + "##### 2.6.4.2.3 fastEight" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DtresAlyTNrn" + }, + "source": [ + "Look at the different fastEight values more closely:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "jNXA8HhiTNrn", + "outputId": "3d09eb90-22bd-461d-fb56-976ba35df3af" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "0.0 163\n", + "1.0 1\n", + "Name: fastEight, dtype: int64" + ] + }, + "execution_count": 34, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data.fastEight.value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "OjGO8QxDTNro" + }, + "source": [ + "Drop the fastEight column in its entirety; half the values are missing and all but the others are the value zero. There is essentially no information in this column." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "PD_sXpbtTNro" + }, + "outputs": [], + "source": [ + "#Code task 24#\n", + "#Drop the 'fastEight' column from ski_data. Use inplace=True\n", + "ski_data.drop(columns=___, inplace=___)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MM0yrCEqTNro" + }, + "source": [ + "What about yearsOpen? How many resorts have purportedly been open for more than 100 years?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "GHUbEcfBTNro" + }, + "outputs": [], + "source": [ + "#Code task 25#\n", + "#Filter the 'yearsOpen' column for values greater than 100\n", + "ski_data.___[ski_data.___ > ___]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "2ia8cEHOTNro" + }, + "source": [ + "Okay, one seems to have been open for 104 years. But beyond that, one is down as having been open for 2019 years. This is wrong! What shall you do about this?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GgwcPRJITNrp" + }, + "source": [ + "What does the distribution of yearsOpen look like if you exclude just the obviously wrong one?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qgB5BNWsTNrp" + }, + "outputs": [], + "source": [ + "#Code task 26#\n", + "#Call the hist method on 'yearsOpen' after filtering for values under 1000\n", + "#Pass the argument bins=30 to hist(), but feel free to explore other values\n", + "ski_data.___[ski_data.___ < ___].hist(___)\n", + "plt.xlabel('Years open')\n", + "plt.ylabel('Count')\n", + "plt.title('Distribution of years open excluding 2019');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ynZBUVW3TNrp" + }, + "source": [ + "The above distribution of years seems entirely plausible, including the 104 year value. You can certainly state that no resort will have been open for 2019 years! It likely means the resort opened in 2019. It could also mean the resort is due to open in 2019. You don't know when these data were gathered!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "AdppyA_3TNrp" + }, + "source": [ + "Let's review the summary statistics for the years under 1000." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "nVyd-8w6TNrp", + "outputId": "35ee0ba2-f55f-4ef0-9bcc-0773a14b4294" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "count 328.000000\n", + "mean 57.695122\n", + "std 16.841182\n", + "min 6.000000\n", + "25% 50.000000\n", + "50% 58.000000\n", + "75% 68.250000\n", + "max 104.000000\n", + "Name: yearsOpen, dtype: float64" + ] + }, + "execution_count": 38, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data.yearsOpen[ski_data.yearsOpen < 1000].describe()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "de5kdbwkTNrp" + }, + "source": [ + "The smallest number of years open otherwise is 6. You can't be sure whether this resort in question has been open zero years or one year and even whether the numbers are projections or actual. In any case, you would be adding a new youngest resort so it feels best to simply drop this row." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "EaDzE9VNTNrq" + }, + "outputs": [], + "source": [ + "ski_data = ski_data[ski_data.yearsOpen < 1000]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "SZ6y5M9JTNrq" + }, + "source": [ + "##### 2.6.4.2.4 fastSixes and Trams" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hPaZqHh2TNrq" + }, + "source": [ + "The other features you had mild concern over, you will not investigate further. Perhaps take some care when using these features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QMMST7PUTNrq" + }, + "source": [ + "## 2.7 Derive State-wide Summary Statistics For Our Market Segment" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Z8PkS6bWTNrq" + }, + "source": [ + "You have, by this point removed one row, but it was for a resort that may not have opened yet, or perhaps in its first season. Using your business knowledge, you know that state-wide supply and demand of certain skiing resources may well factor into pricing strategies. Does a resort dominate the available night skiing in a state? Or does it account for a large proportion of the total skiable terrain or days open?\n", + "\n", + "If you want to add any features to your data that captures the state-wide market size, you should do this now, before dropping any more rows. In the next section, you'll drop rows with missing price information. Although you don't know what those resorts charge for their tickets, you do know the resorts exists and have been open for at least six years. Thus, you'll now calculate some state-wide summary statistics for later use." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jdLv7VpBTNrq" + }, + "source": [ + "Many features in your data pertain to chairlifts, that is for getting people around each resort. These aren't relevant, nor are the features relating to altitudes. Features that you may be interested in are:\n", + "\n", + "* TerrainParks\n", + "* SkiableTerrain_ac\n", + "* daysOpenLastYear\n", + "* NightSkiing_ac\n", + "\n", + "When you think about it, these are features it makes sense to sum: the total number of terrain parks, the total skiable area, the total number of days open, and the total area available for night skiing. You might consider the total number of ski runs, but understand that the skiable area is more informative than just a number of runs." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xEXIlXfcTNrr" + }, + "source": [ + "A fairly new groupby behaviour is [named aggregation](https://pandas-docs.github.io/pandas-docs-travis/whatsnew/v0.25.0.html). This allows us to clearly perform the aggregations you want whilst also creating informative output column names." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "5muKYVSATNrr" + }, + "outputs": [], + "source": [ + "#Code task 27#\n", + "#Add named aggregations for the sum of 'daysOpenLastYear', 'TerrainParks', and 'NightSkiing_ac'\n", + "#call them 'state_total_days_open', 'state_total_terrain_parks', and 'state_total_nightskiing_ac',\n", + "#respectively\n", + "#Finally, add a call to the reset_index() method (we recommend you experiment with and without this to see\n", + "#what it does)\n", + "state_summary = ski_data.groupby('state').agg(\n", + " resorts_per_state=pd.NamedAgg(column='Name', aggfunc='size'), #could pick any column here\n", + " state_total_skiable_area_ac=pd.NamedAgg(column='SkiableTerrain_ac', aggfunc='sum'),\n", + " state_total_days_open=pd.NamedAgg(column=__, aggfunc='sum'),\n", + " ___=pd.NamedAgg(column=___, aggfunc=___),\n", + " ___=pd.NamedAgg(column=___, aggfunc=___)\n", + ").___\n", + "state_summary.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9pI-IqLMTNrr" + }, + "source": [ + "## 2.8 Drop Rows With No Price Data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-NlXdERmTNrr" + }, + "source": [ + "You know there are two columns that refer to price: 'AdultWeekend' and 'AdultWeekday'. You can calculate the number of price values missing per row. This will obviously have to be either 0, 1, or 2, where 0 denotes no price values are missing and 2 denotes that both are missing." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Tjy2GyVLTNrr", + "outputId": "3702c49e-bf8f-4a45-91bf-1880e1e921d2" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "0 82.317073\n", + "2 14.329268\n", + "1 3.353659\n", + "dtype: float64" + ] + }, + "execution_count": 41, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "missing_price = ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum(axis=1)\n", + "missing_price.value_counts()/len(missing_price) * 100" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "w0rkdN-pTNrr" + }, + "source": [ + "About 14% of the rows have no price data. As the price is your target, these rows are of no use. Time to lose them." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Ocm7TbzDTNrs" + }, + "outputs": [], + "source": [ + "#Code task 28#\n", + "#Use `missing_price` to remove rows from ski_data where both price values are missing\n", + "ski_data = ski_data[___ != 2]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3j-ylXeuTNrs" + }, + "source": [ + "## 2.9 Review distributions" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "uFNrshckTNrs", + "outputId": "adc97012-e084-44cc-ab5c-e50fd299e880" + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA24AAAJOCAYAAADPilTlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebwcVZn/8c+XhCWEsEYjSySKiIIosinjFsUFxBEcRUFUMi7oKKOMUQH1p+iIRkccXHBlVXYXIIMLInJBRPYt7AQIJCQQwpoEVALP749zOqnb6b63+95eqm9/369Xv253VXXVU3WfPn3OqVPVigjMzMzMzMysvNbodgBmZmZmZmY2NDfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NtzoknSjpa61etp2ajHmGpEvaHZN13ljP3VFuZ0DSh9u9HbN2kvR7SQd2O45+J2mZpOc3uGxIekGLtjstr298jXnPzXGNy6+dK2ZjSF823HLl7RFJa7dh3dMlLSi83jQXsFMK075QZ9ofWh2PjS3OXesVkuZJemO34xhKdc7naUdIeipXfh+VdKmk3boVYy0RsWdEnNTtOMa6nMMPSJpYmPZhSQMAEbFeRNzVgu2s1pEqaQtJv5a0RNJjkuZImjHcuiLi3hzX0/m1c8U6Jn9mnszl5/25U3a9bsc1lvRdw03SNOA1QABvb/f2ImIRMBd4bWHya4Fba0y7uN3xWO9y7pp1zBkRsR4wGbgQ+GWX47HuGQ98qgvb/QUwH9gS2AT4APBAF+Iwa9a/5vJzB+DlwOFdjmdM6buGG6nwuww4EVg5fEDSyyVdI2mppDOAdQrzavWGrTbsIffK/R7YLPc2LJO0GalS+9q8zDhSIn+3atpueTkkfVDSLfnMynmStixs40WSzpf0sKTbJL271k5KmiTpQknfU7KJpNmSHpd0BbBV1fLflTQ/z79a0mvy9OdIekLSJoVld5L0oKQ1Gzje1jr9mrt135d7846R9Nu8/5dL2qow/02Sbs091j8A1PDRtpaTtLakoyUtzI+jlc8eV85+SZopabGkRZL+vfDeTST9Xy6jrpT0tWJuD5Mnb5V0c86R+yR9ZoicXykiVgCnAJtLelZe16AziUpn6E7OzytD2A6UdK/S2ZIvNHBcjpD0S0kn5xjnSHqhpMPzsZgv6c2F5T3kt3P+B/iMpA2rZxTL0uHyM3ujpDty+XhMLt9eDPwY2C3n4KN52V2AEyNieUSsiIhrI+L3tQKU9M6cly9R1TDKYq5Uvg8kfTvHcLekPQvreZ6ki3MO/inHePJwByjn7v25nL1Y0naFeRMkHSXpnjz/EkkThlun9b6IuB84D9hBtUc3rCxLcxl4pqSf5/y7SdLOhWUPzWX30ly+797ZvSmPfm24nZIfb5E0RdJawNmkHq6NSb2r72x2xRGxHNgTWJiHKqwXEQspVH5JFd9bgQuqpq0JXCFpH+DzwL8BzwL+ApwGKyvX5wOnAs8G9gd+WCwk83Kb5PX/NSI+GREBHAP8HdgU+GB+FF1J6h3ZOK//l5LWyR+8AaBYyX4fcHpEPNXsMbJR6bvcBdZt4H37A18BNiKdITwyr2sy8Gvgi6QzJ3cCr2r22FhLfQF4JamseRmwK+n/U/EcYANgc+BDwDGSNsrzjgGW52UOZHDnxXD5dRzw0YiYBLwE+PMQOU9hvWuRPncPAY80sZ+vBrYBdge+lCvnw/lX0ud4I+BaUoVnjXwsvgr8pIntW+tcRfoO/Mwwy9XNz4K3kRpkLyN9p74lIm4BPgb8LedgpYF4GSn/95P03HobVerc+Cbwxoi4sYH9eQVwG6lM/BZwnKRKh9apwBWkM3xHAO9vYH2QOkC2Jn32riF9R1V8G9gJ+BfSd9TngGcaXK/1MElbkMrYuQ2+5e3A6cCGwGzgB3k92wAHA7vkMvwtwLxWx9sr+qrhJunVpGEHZ0bE1aSK3HtJFYk1gaMj4qmI+BWpIdMqFwEvyRWQ1wB/iYg7gMmFaZdFxD+BjwLfiIhbcm/v10m9FVuSCv15EXFC7oG7hlQxfVdhW5vl7f0yIr6Y93scqTL/pdx7dyMwaMx7RJwcEQ/l9R4FrE2qeJCXfV9hXfuTKhjWIf2auw2+7zcRcUXh7MgOefpbgZsj4le5k+Fo4P4WHhtr3gHAVyNicUQ8SGpwFyuHT+X5T0XE74BlwDaFMuzLEfFERNzM4DJsuDx5CthW0voR8UieP5R35zMfTwIfAd6V86tRX4mIJyPieuB6UkV9OH+JiPPydn5J6vyYlXP3dGBarbM+1hFfAv5T+axrtQbys2JWRDwaEfeShuDuUGOZin1JnV//D7hb0nWSdqla5hDgs8D0iGi0cnxPRPwsXwN3Eqkzd0puHO5Cqif8MyIuIVWehxURx0fE0oj4B6nB9zJJG0hag9RJ/KmIuC8ino6IS/NyNnadLWkpaajvYuDLDb7vkoj4Xc7NX7Cq3HyaVCfdVtKaETEvIu5sedQ9oq8abqQesD9GxJL8+tQ8bTPgvnxmquKeVm00IuYBC0i9sK8lFcYAfytMq1wjtCXwXaWL4h8FHiYN79o8z3tFZV6efwCph69iL2ACaehFxbNI4/Tn19s/peFJt+ShDI+Ser0n59nnkD4wzwfeBDwWEVeM6GDYSPVr7jbyvmJj7AmgciH0ZhRyPh+j4mfAOm8zBufnPXlaxUNVDaTK/7NWGVZ8PlyevJPUkL9H0kUa/mYjZ+YzH1OAG0lnDJpRLyeHUrx+6UlgSeUGE/k1Da7HWix3dp4LHFZnkeHys6LhvMgdDIdFxHakPLyOVCEuDvf+LHBMRCyouZLaVsYQEU/kp+uRPocPF6bV24dBJI2TNEvSnZIeZ9WZkMn5sQ6po9H6xz75zNh04EWsqksOp/rzsY6k8blT4hBSp8BiSaeralh7P+mbhlseU/1u4HV5LPb9wH+RWvSLSNcwFAvE4tCE5aQhW5V1FSuN1aLO9L+QKrm7AZdWTXs1qyq/80lDejYsPCZExKV53kVV89aLiP8obOdnwB+A32nVnbAeBFYAU2vtn9L1bIeSjs9GucLyGPl6oIj4O3AmqSL0fny2raP6PHcbeV89iyjkfD5GU+svbh2wkNTIqnhunjacShm2RWFa8X85ZJ5ExJURsTdpKNfZpPIM6uc8+X1LSGeSj5C0aZ486DPF4E4EG7u+TDr7unmNecPl53AaycNvkxpXGxdmvRn4oqSmh8fXsAjYWFIxtxvZh/cCewNvJHX4TsvTBSwhXaKxVc132pgWEReRrsn/NqvXRcaROjwaXdepEVEZeRSk4cF9qW8absA+pNOt25KGJ+wAvJhUAd2HVOh+UtJ4Sf9Guvai4npgO0k7SFqH1Oqv5wFgE0kbVE2/mHStxMKIeDxPuyRP24B0BgPS2YbDK9dm5OEG++Z55wIvlPR+SWvmxy41rp84mDSG/VxJE3Kv7W9IlY91JW3L4PH3k/L+PwiMl/QlYP2qdf4cmEEagzzsxcrWUn2bu028r5bf5n3/N6UL9T+JK9mdtqakdSoP0jWPX5T0LKVrEL9EA+VJjTLsRaT8q6ibJ5LWknSApA3ysMPHSZ8nqJ/zxW3fSrre7HN50nXAfnkbOzN42K6NUbnX/wxSOVI9b7j8HM4DwBZK11QCIOmbSjcbGS9pEvAfwNyIeKjwvpuAPUjXwo3qTsMRcQ/per4j8mdmN9J1l8OZBPyDdB3ouqQh8pV1PgMcD3xH0mb57NxuasPP2VhpHU0aqbUu6QzaXko3tvsiafjjsCRtI+kNOW/+ThqB8PQwbxuz+qnhdiBwQqTfOLm/8iBd/Lg/6YYKM0gXoL+HVAgDEBG3ky4O/xNwB6nSWlP+kj8NuCsP2amczr2I1NtbfO91pKFhV1eGJ0TEWaSehNPzsIMbSRd3EhFLST1s+5F6qe/Pyw5K/jwk7CBSL/Q5ucJ0MGk4xP2kHpATCm85j3Rx8e2koUt/p2qIRET8lXRB8TV5+Jx1Tt/mLunapGHfV2d/lpCuE5lFqlRsDfx1uPdZS/2O9CVbeaxDqhzeAMwh3cig0R9eP5jUUXA/6az/aaQKYyP59X5gXs7Lj5Gv2R0i56v9D3CQpGeTrjnaivR5+wpp2LL1h68CE+vMq5ufDfgzqRF2v6TKcPh1gbOAR4G7SGcaVmuc5eso3wb8TIU7RI7QAaSRFQ+RPpdnMPw+/JxUb7gPuJl0U5Wiz5A+61eShs9/k/6qe/a1SNcy/5zU4fFx4FhSriwnXYbRiLVJ3+NLSJ+vZ5NuhNaXNPjSGLP6JP0ZODUiju12LGbW3yR9E3hORNS6e59ZV42F/FT6eZlbI6LRm0uYWZu518MaonQ3qx1JPXBmZh2l9DttL1WyK+nnAs7qdlxmMDbyMw8v3krSGpL2IF27dna34zKzVdxws2FJOok01O6QPCTJzKzTJpGGAS8n3VzkKNJw2p4g6fda9UPfxUffDvkZY3o6P7PnkH6zbhnwPeA/IuLafI1ordy9qavRmvUhD5U0MzMzMzMrOZ9xMzMzMzMzK7nx3Q4AYPLkyTFt2rSuxrB8+XImTqx3s6jucExw9dVXL4mIhn/roxu6mb9lzBFwXBVlz996uVvW/1+RY2yNejGWPXeht/K3jDHB2I2r7PlbK3fL+L9wTI1pZUzD5m5EdP2x0047RbddeOGF3Q5hNY4pArgqSpCjQz26mb9lzJEIx1VR9vytl7tl/f8VOcbWqBdj2XM3eix/yxhTxNiNq+z5Wyt3y/i/cEyNaWVMw+Wuh0qamZmZmZmVnBtuZmZmZmZmJVeKa9xaadphv21q+Xmz9mpTJGbNaTZ3AU7co1zjvK1/uey1sphz32PMaCIfnYtWFiOpBzh/+4vPuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZyY25a9yaVRlPPHP7FQ2NifdYYjMzMzMz67S+b7iZmdlgzd7cwczMzNrPQyXNzMzMzMxKzg03MzMzMzOzknPDzczMzMzMrOR8jZtZD/MPzZqZmZn1B59xMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzEpufLcDMLPOmXbYb5taft6svdoUiZmZmZk1w2fczMzMzMzMSs4NNzMzMzMzs5Ib1VBJSfOApcDTwIqI2FnSxsAZwDRgHvDuiHhkdGGamVmFy14zM7P+04pr3F4fEUsKrw8DLoiIWZIOy68PbcF2zMxsFZe91pPK1vHQ7LW/4Ot/xzJJxwNvAxZHxEvytLr5Kelw4EOkfP5kRJzXhbCtT7RjqOTewEn5+UnAPm3YhpmZDeay13rJ6yNih4jYOb+udDxsDVyQX5t1w4nAHlXTauanpG2B/YDt8nt+KGlc50K1fjPaM24B/FFSAD+JiJ8CUyJiEUBELJL07FpvlHQQcBDAlClTGBgYGGUoycztV4zofVMmNPbeVsXZiGXLlnV0e40oY0wjVbZeX7MmtLXsbbQ8HI3RliO9UBY5xqbsDUzPz08CBvAZY+uCiLhY0rSqyfXyc2/g9Ij4B3C3pLnArsDfOhGr9Z/RNtxeFRELcwXhfEm3NvrGXNH4KcDOO+8c06dPH2UoyYwRDHmAVEk5ak4Dh2PO8qbXPdIhFQMDA7TquLRKGWMaJQ83s17U1rL3+6ec01h5OArzDlh9u83ohbLIMdbVdx0PJWogD+K4GlYvPzcHLisstyBPW81wubts2TJmbv9004G18ziV8P/Q9zGN6ps5Ihbmv4slnUXqZXhA0qY5sTcFFrcgTrNOca+vld5YKHv9m4J9re86HsraiHdco6Ya06LWgsPl7sDAAEddMoKTA6PsBBtKGf8P/R7TiEs2SROBNSJiaX7+ZuCrwGzgQGBW/ntOKwI1a4NSDfUdSQ9xu3uWR7pfZewRg/LG1QyXvdbrxkLHg/Wdevm5AJhaWG4LYGHHo7O+MZouqSnAWZIq6zk1Iv4g6UrgTEkfAu4F9h19mGZtUaqhviMZ5tvwEN8RGmlPXhl7xKC8cTXJZa/1LHc8WI+ql5+zgVMlfQfYDNgauKIrEVpfGHGNLyLuAl5WY/pDwO6jCcqsE9zra73IZa/1OHc8WKlJOo10ycRkSQuAL5MabKvlZ0TcJOlM4GZgBfCJiGj+QjWzBrV3ELhZSbnX18ys89zxYGUXEfvXmVUzPyPiSODI9kVktoobbtav3OtrZmZmZj3DDTfrS+71NTMzM7Neska3AzAzMzMzM7OhueFmZmZmZmZWch4qaWZmZmbWg6Y1+VNC82bt1aZIrBN8xs3MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzkfFdJMzMrveo7p83cfgUzhrmbmu+eZmZmY4nPuJmZmZmZmZVcqc+4NfvbFGZmZmZmZmORz7iZmZmZmZmVXKnPuJlZdzV71tvXFJmZmZm1h8+4mZmZmZmZlZwbbmZmZmZmZiXnhpuZmZmZmVnJ+Rq3DvB1QmZmZmZmNho+42ZmZmZmZlZybriZmZmZmZmVnBtuZmZmZmZmJeeGm5mZmZmZWcn55iRmZmZmTWj2pmMn7jGxTZGYWT9xw83MzMakZivXzfIdgM3MrJM8VNLMzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzkfFdJszZp9x3tzMzMzKx/uOFmZmZmZtYHmulUnrn9Cqa3LxQbAQ+VNDMzMzMzKzmfcTMzMzNrozn3PcaMJs50+MfdzawWN9xKqHIae+b2Kxoq6F3AW1k0m7vg/DUzMzNrhBtuY0CzN8FwRdnKxPlrZmZWTiO50Zq/p9vH17iZmZmZmZmVnBtuZmZmZmZmJde2oZKS9gC+C4wDjo2IWe3allkrOXetlzl/rVc5d62XOX9X8SUQ7dOWhpukccAxwJuABcCVkmZHxM3t2J41ZzQ/DN3MTSeaUZYPrXO3/Jy/9Tl/O2u4XGxXvrXSiXtM7HYIgHO32lio+HairC7Lfjt/rVPadcZtV2BuRNwFIOl0YG/ACWxl59y1Xub8tV7l3O2waYf91ncAbh3n7yiMhR8F71RniyJiRG8ccqXSu4A9IuLD+fX7gVdExMGFZQ4CDsovtwFua3kgzZkMLOlyDNUcE2wZEc/q1MYayd08vSz5W8YcAcdVUbr8bTB3y/r/K3KMrVEvxtLlbp7eq/lbxphg7MZVuvxtIHfL+L9wTI1pZUxD5m67zripxrRBLcSI+Cnw0zZtv2mSroqInbsdR5Fj6ophcxfKk79l/X84rq5pSdnbC8fJMbZGiWJsWdlbon1aqYwxgeNqoVGXvWXcZ8fUmE7G1K67Si4AphZebwEsbNO2zFrJuWu9zPlrvcq5a73M+Wsd0a6G25XA1pKeJ2ktYD9gdpu2ZdZKzl3rZc5f61XOXetlzl/riLYMlYyIFZIOBs4j3Rb1+Ii4qR3baqGuD3urwTF1WA/mbln/H46rC1qYv71wnBxja5QixhaXvaXYpypljAkcV0u0KH/LuM+OqTEdi6ktNycxMzMzMzOz1mnXUEkzMzMzMzNrETfczMzMzMzMSq4vG26S5kmaI+k6SVflaRtLOl/SHfnvRh2I43hJiyXdWJhWNw5Jh0uaK+k2SW/pYExHSLovH6/rJL21kzFZ4rwdUVzO3SZJ2iMfk7mSDut2PNUkTZV0oaRbJN0k6VPdjqkeSeMkXSvp3G7HUoukDSX9StKt+Xju1u2YRqObuVsvL0tSNg7Kw5LEtFrulSGubulW7pbt+7yMnyNJ60i6QtL1OaavdDWmiOi7BzAPmFw17VvAYfn5YcA3OxDHa4EdgRuHiwPYFrgeWBt4HnAnMK5DMR0BfKbGsh2JyY+Vx9t523xczt3mjuG4fCyeD6yVj9G23Y6rKsZNgR3z80nA7WWLsRDrp4FTgXO7HUud+E4CPpyfrwVs2O2YRrEvXc3denlZkrJxUB6WJKbVcq8McfVb7pbt+7yMnyPSb/Stl5+vCVwOvLJbMfXlGbc69iYVJOS/+7R7gxFxMfBwg3HsDZweEf+IiLuBucCuHYqpno7EZENy3g4dVz3O3dp2BeZGxF0R8U/gdNKxKo2IWBQR1+TnS4FbgM27G9XqJG0B7AUc2+1YapG0PqnSdhxARPwzIh7tblSj0tXcHSIvu1o21snDbsdUL/e6/j3SJV3L3bJ9n5fxcxTJsvxyzfyIbsXUrw23AP4o6WpJB+VpUyJiEaTEAZ7dpdjqxbE5ML+w3AI6W1k5WNIN+bR65XRwt2PqN87bkXHuNq6njoukacDLST2gZXM08DngmW4HUsfzgQeBE/IwumMlTex2UKNQmtytystul4218rDbMdXLvW7H1S1l279S/B/K9DnKw42vAxYD50dE12Lq14bbqyJiR2BP4BOSXtvtgBqgGtM69VsOPwK2AnYAFgFHlSCmfuS8bZ5ztzk9c1wkrQf8GjgkIh7vdjxFkt4GLI6Iq7sdyxDGk4ZI/SgiXg4sJw336VWlyN0m8rLt8Y4gDzt1DJvNvVL8b9uoV/avY3GW6XMEEBFPR8QOwBbArpJe0q2Y+rLhFhEL89/FwFmkU5gPSNoUIP9d3KXw6sWxAJhaWG4LYGEnAoqIB3LSPgP8jFWnfLsWUz9y3jbPudu0njguktYkfamfEhG/6XY8NbwKeLukeaRhT2+QdHJ3Q1rNAmBB7jkG+BWpMt2rup67dfKym2VjvTzsdnldL/e6HVe3lG3/uvp/KOHnaKU8pHcA2KNbMfVdw03SREmTKs+BNwM3ArOBA/NiBwLndCfCunHMBvaTtLak5wFbA1d0IqBKYmbvIB2vrsbUb5y3I+PcbdqVwNaSnidpLWA/0rEqDUkiXRtzS0R8p9vx1BIRh0fEFhExjXQM/xwR7+tyWINExP3AfEnb5Em7Azd3MaTR6mruDpGXXSsbh8jDrpbXQ+ReKb9HOqBs5W7X/g9l/BxJepakDfPzCcAbgVu7FlOr7nLSKw/S2Orr8+Mm4At5+ibABcAd+e/GHYjlNNLwradILfQPDRUH8AXS3WluA/bsYEy/AOYAN+SE3LSTMfnhvB1FXM7d5o/jW0l38rqzkmdlegCvJg07uQG4Lj/e2u24hoh3OuW9q+QOwFX5WJ4NbNTtmEa5P13L3Xp5WYayMW9rZR6WIaZauVeGuLr16Fbulu37vIyfI+ClwLU5phuBL+XpXYlJeQNmZmZmZmZWUn03VNLMzMzMzKzXuOFmZmZmZmZWcm64mZmZmZmZlZwbbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcm64mZmZmZmZlZwbbmbWdpJC0gvqzFsm6fmdjsn6l6SbJE3vwnYHJH2409s16wRJ75A0P5fpLx9m2RmSLim8rvsdYWarjPmGW3XhUDXvAEl/bHA9R0g6eYj58yS9caRxtpuk5+bCdFy3Y7H6JL1a0qWSHpP0sKS/Stqli/EM5C/Ul1VNPztPnz7abUTEehFx12jXY70ll0eVxzOSniy8PqCd246I7SJioME45xVie0DSCZLWa2d8Zs2StJakoyQtyLl6t6T/7XAY3wYOzmX6tR3etrVI2euzAJJOlPS1bsfRDWOm4TaSCm9EnBIRb+5gjK8pVEyW54pvsfLy3HZtOyLuzYXp0+3aho2OpPWBc4HvAxsDmwNfAf7RzbiA24EPVF5I2gR4JfBg1yKynpfLo/UiYj3gXuBfC9NOaWQdksY3Mq0F/jXHuSOwC/DFZt6sZMx831rz2pSXRYcDOwO7ApOA1wOdbjxtCdzU4W2a9ZUx8UVS4grvIBHxl0JFZbs8ecNCZeXeRtZTp7LiM2m974UAEXFaRDwdEU9GxB8j4gZYdfZY0rclPZJ7VPesvFnSZpJm546LuZI+kqevk88YTM6vvyhpRf7cIOlrko4eIq5TgPcUcmx/4Czgn4Vt7yrpb5IelbRI0g8krVVrZbmTZb6k1+fXK4fI5F60YyT9VtJSSZdL2qrw3jdLui130PxQ0kXDDT2TtJWkP0t6SNISSadI2rAwf6qk30h6MC/zg6HWZ+0laQ1Jh0m6M/8/zpS0cZ43LefLhyTdC/w5fy7+Kul/JT0MHNHA/3xlj7LSaIozJf0859xNknauFVtE3Af8HniJpI0knZvz5pH8fIvCNgYkHSnpr8ATwKDhwJI2lXSDpM/k1zMk3ZVjuFttPutogxVybqmkm5WG/a2dy7SXFJZ7Vi5Pn51fv03SdXm5SyW9tLDsPEmHSroBWC5pfK3tFJYfp3TWbEnOgYNzvo/P8zeQdFwuY+/LZXelXN4FOCsiFkYyLyJ+XhXLZ3LOPSbpDEnrFOZ/ROl742Gl75HN8vSvSPp+fr6mUqfzt/LrCZL+LmmKpGXAOOB6SXfWO6Yt/reZ9Z0x0XBjmApvkaT/Uar8bqDVx1h/V6lC+bikqyW9purt6+TCbqmka1Q1fKywnroVj3qGKpDrVExOlPQjSb+TtBx4vaS9JF2b458v6YjC+isVnsoXwICk/87rXSrpj8oV+2Hi/KWk+3PBf7Gk7QrzJuQvnXvy/EskTRhunbbS7cDTkk6StKekjWos8wrgNmAy8C3gOEnK804DFgCbAe8Cvi5p94j4O3Al8Lq83GuBe4BXFV5fNERcC4GbgcrZ6Q8AP69a5mngv3JcuwG7Ax+vXpGkt+Q43xkRF9bZ3v6kjpeNgLnAkfm9k4FfkXqWN8nH4V+GiHvlZoFvkI7Li4GpwBF5neNInT73ANNInT6nN7BOa59PAvuQ8nUz4BHgmKplXkf6X74lv34FcBfwbFK+1P2f1/F20v99Q2A2ULPxLmkq8FbSmYw1gBNIZxmeCzxZ433vBw4inQG5p7CeaaTP3A8i4tuSJgLfA/aMiEmkvL5uiHit9e4EXgNsQCp/TiZ1BP+GVCZVvBu4KCIWS9oROB74KKlM+gkwW9LaheX3B/YiddKuqLUdSZvmZT8C7AnsQDq7u09VjCcBK4AXAONkdEgAACAASURBVC8nlcmVjqvLgE9L+rik7QvfC0XvBvYAnge8FJgBIOkNpM/Lu4FNSblaKQcvAqbn57sA97Pqu2Q34LaIeCB3SAO8LCIqnW1D7auV2y65sf2I0vDwdTR8Z1XdzidJH5R0S37feZK2HC4ASS+SdH7uTLhN0ruHWLZmB0qui/+qatnvSvreyA5LCUREzz+A9YGHSIXansBGhXkzgEtIX7I/A84D1i3OKyz7PlLhOx6YSSqg1snzjgCeIlWI1wQ+A9wNrJnnzwPemJ8fQipEtwDWJhXmp1XFPA0IYHx+fXZebiKp8nEF8NFCnCuA/8yxTQBOBB4jVb7XANYhFa7b59cvBR4A9qmzvQFSofrCvL4BYFYDx/qDpErI2sDRwHWFecfk9WxO6nn7F2DtbudHLz1IlcwTSQ2wFaRK5JRCHswtLLtu/p8+h1QxfRqYVJj/DeDE/Py/SRXD8TmvPwXMynnzJDC5TjwDpIrB+0gNrm2A2/O8BcD0Ou87hNT7W3kdpAbXPcD2VcsG8IL8/ETg2MK8twK35ucfAP5WmCdgPvDhJo/xPsC1+flupCGf47v9v+/nB4PLz1uA3QvzNiWVveML5djzC/NnAPc2+j+vsb0jgD8V5m0LPFm17DLg0Zy/PwQm1NjGDsAjhdcDwFerlhkAvpPXuX9h+sS8/nfWWrcfXcnJ64C9gTcCdxWm/xX4QH7+I+C/q953G/C6Qu58sJHt5Od/Jn/v59dvzPk+HphCGkU0oTB/f+DC/Hwc8Ikc3z9IHW4HFpadB7yv8PpbwI/z8+OAbxXmrZc/c9NI9YO/k+pGhwGfJ5X965EaY98rvG9lWd7Avs5gcP1ryPf60dHcnwfcSKpXbJxz6ms5B95JqntMAn4JnJ3fMxF4HNgmv94U2C4/34fUCfvinMtfBC4dJoaJpO/3f8/v2RFYUljnicDX8vMdgcWkDrxxwIF5H9Ymdaw9Aawfqz4ni4BXdvs4j/QxJs64RcTjwKtJH/yfAQ8qneqfkhdZk1Tp3Jh0rcITddZzckQ8FBErIuIo0j99m8IiV0fEryLiKdKX7zqka32qfRT4QkQsiIh/kCoG71KdMe45zj2BQyJieUQsBv4X2K+w2MKI+H6O7ck87ZyI+GtEPBMRf4+IgYiYk1/fkPf5ddR3QkTcntd3JqniMaSIOD4ilhb262X5bOEapEbdpyLivkhnPi/Ny1mDIuKWiJgREVsALyGdMSgOY7y/sGwlj9fLyz0cEUsLy95DakTDql7THYE5wPmk3HglqTG4ZJjQfgO8gdR58IvqmZJemHvf7pf0OPB10tm3okOAMyNizjDbur/w/Im8f5D2cX5lRqRSeMEw60LSsyWdns9kP07qSa/ENhW4J1JPuJXDlsBZuef0UVJD7mlSxbViftV7Br0e5n9eS3XOrVNVXu8TERtGxJYR8fGIeFLSupJ+kkcYPA5cDGyowcPWq+MEOAC4j3T2GICIWA68B/gYsEhpqPCLhojXWkzSBwo99o+Syt/JpMbUBEmvyGcJdiANFYeUqzMr78nvm0oqqyqqc7PedqCqjKt6viWpLrOo8N6fkDp6yd+5x0TEq0hnjo8Ejpf04sI6hipbV54RjohlpM7wzXP94CrS90VldMalpE7j1zHEaI1h9tXK7QcRMT8iHibl0v65fvzriHgi1zWOZHAd8xnSMPIJEbEoIirXO34U+Eau36wg1Q92GOas29uAeRFxQq73XgP8mnTypNpHgJ9ExOX5c3ASqfPilRFxD3ANq85evwF4IiIuG8lBKYMx0XCDYSu8LyD1nH0lIv5Zbx2SZuZTuY/lQmYDBhcyxUrjM6wallatkYpH9fJ1C+Tqbdeblr9YLsynsR8jVQKaqawMeac0pfH3s5SGgD5O6tEgb2MyqSF751DrsMZFxK2kXqWXDLMopN7VjSVNKkx7LqmCCOmLdhvgHaRhPjfn+Xsx9DDJSixPkK7t+Q9qNNxIPc+3AltHxPqkXtnqoTr7AvtIOqSB/allEeksNpBu+FB8PYRvkDp1Xppje18htvnAc+t1qlhXzCcNGdyw8Fgn0vVlFVH1nurXQ/3PW2Um6TP1iryN1+bpxe1UxwWpw2sJcGqxkRcR50XEm0g91beSOiGtA3IF8mfAwcAmEbEh6YyD8nf9maSzW+8Fzi10kM0HjqzK1XUj4rTC6qOR7eRFBpVxpEZgxXxSZXRyYVvrR8R2VIl0ucgxpGHG2zZwCBaS6iGVOCeSzq5UPnMXkSq8LycNu7+INEx5V1KHxWoa2Fcrt2L98h5gs6E6q4bpfNoS+G6hfvswKQ82p74tgVdUdYocQBphVGvZoTpQTmXVcOf35tc9a8w03IpqVHhvIZ1u/b2kbWq9R+l6tkNJY7w3yoXMYwwuZKYWll+DVMAurLG6Rioe1csPVyDXqgBUTzuVNLRuakRsAPyY1haS72XV0JENSMMoyNtYQhpOsVXNd9qw8njumZUx40rX0+xPGnY7pIiYT2qcfUNpLPpLgQ+RbixSaXhdTRpKU2moXUrqCRu24ZZ9njQEaF6NeZNIwySW5cL6P2oss5B07dsnJa12/VsDfgtsL2mf3ND6BLUL8VqxLQMelbQ58NnCvCtIlaVZkibmY/eqWiuxjvkxcGSlN1bpZhB7N7mOof7nrTKJNMz4UaVrmL/c4PueInViTAR+oXRN9BRJb88V5n/k2H0H4M6ZSPo+fRBA0r8zuMPsVFKl9AAGV/p+Bnwsd5oqlyF7VXWgNbOdM4FPSdpc6WY6h1ZmRMQi4I/AUZLWz3mzlaTX5XUdImm60rXm4yUdSMrRRu4seSrw75J2ULo+7+vA5YWy/iLSUPWbc+f3AGkI/d0RUe/uwsPtq5VbsdPguaTv7yE7q4bofJpPGgJcrBNPiIhLh9j+fFInc/E960VErbrFcB0ovwSm57rVO3DDrfsaqfDmf+DngT+pcJe6gkmka4oeBMZL+hLp2rminST9W640HkL6gq1VqW6q4jFcgdyESaThcn+XtCupodVKk0j7/BBpjPPXKzNyr+TxwHeU7m44TtJuGnyRtg1tKWmM9uVKN5y5jNRDObPB9+9PakwvJA3l+XJEnF+YfxHpzO4VhdeTqNNjWi3S3cpq/iYi6ZrP9+Z9+BlwRp113EtqvB2qJn+IOA/n3Jd0bcZDpJ7kqxj+7rFfIQ0RfYzU+PtNYZ1PA/9KOit/L+ks+nuaicta7rukDqg/SlpK+hy8osl11P2ft9DRpOt/lpBi/EOjb8yV338jjao4nlXXVS8k9Ua/jho397H2yCMQjgL+Rro2fHvSdT2V+ZcDy0k9+L8vTL+KNEzrB6SzW3PJN/wYyXZIZecfgRtIDa7fkeollUb8B4C1SDeLeoQ03LZys48n87rvJ+XkJ0g3gRr2NzIj4gLg/5GGoi0idcAWL9W4lJTrle+Km0kdtXW/OxrYVyu3T0jaIndKfZ70nV63s2qYzqcfA4cr38xO6fKafYfZ/rnACyW9X+lupmtK2kWDh/5WDNmBkjsXBkg3k7o7Im4Z0REpiyjBhXajfZBOt55JOq2/PP/9CanhNYPBF8B+hFV3kFs5j3TB4nGkswaLgM+x+gXsvyIl71JSobpjYb3FZdcAPk26SHkpafjg16tinsbgm4VsQBputoBU2bgW2C/PG7QPUXVhZmHau/K+LSUl/Q+Ak+tsb4DCTR1qbaPGcV4POCev/x7Sl0jxxhITSJWZ+/I+XIwvtPejTY/8OVsIvL7bsfjhhx9+tPpBuvb9nm7H4Ud/PXJ99nBSA/1R0o3/1iV1XAyQGmW3k0bsVG6esympM/ix/J4BYNvCOt9Pur7+cdIZsuMbiGMbUsfbg6TO2j8DO+R5g+rApLulXpm3vYh0lm1S1fYD+Gy3j+9oH8o7ZGZWeko/J3A5qdfvs6Re5efHqhv2mJn1JKWfz3k96azbFNIZsMsiYqTXBZvZGDMmhkqa1SPpeEmLJd1YmLax0m+D3JH/blSYd7jSj5DelhsJVi67kc5gLyENcdwn0h3+fixpWY3Hj7sbrplZw0Qa5vsIadTNLcCXuhqRmZWKz7jZIEo/mPiTGrPuiRp3ryo7Sa8lndb/eUS8JE/7FulawFmSDiPdjOZQSduSfkJhV9KQgD8BL4x0HZSZmZmZtUC+KeDva82LVT/oblXccLMxT9I00i2cKw2320g/HL1I0qbAQERsI+lwgIj4Rl7uPOCIiPhbdyI3MzMzM0tK8dtFkydPjmnTpq02ffny5UycOLHzAQ2hbDGN9XiuvvrqJRHxrJatMJkS6U6e5MZb5ffyNmfwXUIXUOd3RiQdBBwEMGHChJ2mTp262jLPPPMMa6xRjtHIjqW2dsdy++23tyN/W6Ze2dsOZSmrHEdjcbSp7G2pYv6W5XgWlTEm6I+4yp6/tcresv5fWq0f9nM0+zhs7nb77igRwU477RS1XHjhhTWnd1PZYhrr8QBXxejvkDQNuLHw+tGq+Y/kv8cA7ytMP450O+Wez1/HUlu7Y2lF/rbzUS9326Es/3fHMVi9OMqeu1GVv2U5nkVljCmiP+Iqe/7WKnvL+n9ptX7Yz9Hs43C5W45ub7POeiAPkST/XZynL2Dwj07W+4F1MzMzM7OOKsVQyXrm3PcYMw77bVPvmTdrrzZFY2PIbOBAYFb+e05h+qmSvkO6OcnWrPqx6qY1m7/OXSuLaU2Wu+D8tfJoNn+du1YWLnttOKVuuJmNlqTTgOnAZEkLgC+TGmxnSvoQcC+wL0BE3CTpTNKPTq4APhG+o6SZmZmZlYAbbjamRcT+dWbtXmf5I4Ej2xeR2dg0XE/xzO1XDDoD7V5iMzOz5vgaNzMzMzMzs5Jzw83MzMzMDJB0vKTFkm4sTNtY0vmS7sh/NyrMO1zSXEm3SXpLd6K2fuGGm5mZmZlZciKwR9W0w4ALImJr4IL8GknbAvsB2+X3/FDSuM6Fav1m2IabpKmSLpR0i6SbJH0qT3fvg5mZmZmNGRFxMfBw1eS9gZPy85OAfQrTT4+If0TE3cBcYNeOBGp9qZGbk6wAZkbENZImAVdLOh+YQep9mCXpMFLvw6FVvQ+bAX+S9ELfnc/MzMzMetCUiFgEEBGLJD07T98cuKyw3II8bTWSDgIOApgyZQoDAwOD5i9btoyZ2zdfVa5eT9ktW7as52JuVjv3cdiGW07USrIulXQLKSn3Jt1mHVLvwwBwKIXeB+BuSZXeh7+1OngzMzMzsy5RjWlRa8GI+CnwU4Cdd945pk+fPmj+wMAAR12yvOkA5h0wfdhlymRgYIDqfR9r2rmPTf0cgKRpwMuByxll78NwPQ8AUyakW0g3o92t+LL1FDgeMzMzs7Z6QNKmub67KbA4T18ATC0stwWwsOPRWd9ouOEmaT3g18AhEfG4VKuTIS1aY9pqvQ/D9TwAfP+UczhqTnM/Ndfunoey9RQ4HjMzM7O2mg0cCMzKf88pTD9V0ndIlwdtDVzRlQitLzTUKpK0JqnRdkpE/CZPdu+DmZmZmY0Zkk4jXQo0WdIC4MukBtuZkj4E3AvsCxARN0k6E7iZdE+IT/ieDtZOwzbclE6tHQfcEhHfKcxy74OZmZmZjRkRsX+dWbvXWf5I4Mj2RWS2SiNn3F4FvB+YI+m6PO3zuPfBzMzMzMysIxq5q+Ql1L5uDdz7YGZmZmZm1nbD/gC3mZmZmZmZdZcbbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcm64mZmZmZmZlZwbbmZmZmZmZiXXyO+4mZlZiUiaBywFngZWRMTOkjYGzgCmAfOAd0fEI92K0czMzFrLZ9zMzHrT6yNih4jYOb8+DLggIrYGLsivzczMbIxww83MbGzYGzgpPz8J2KeLsZiZmVmLeaikmVnvCeCPkgL4SUT8FJgSEYsAImKRpGfXeqOkg4CDAKZMmcLAwMBqy8zcfkXLA54yYfB6a223E5YtW9a1bTsOMzMbDTfczMx6z6siYmFunJ0v6dZG35gbeT8F2HnnnWP69OmrLTPjsN+2Ks6VZm6/gqPmrPrKmXfA6tvthIGBAWrts+MwM7Oy81BJM7MeExEL89/FwFnArsADkjYFyH8Xdy9CMzMzazU33MzMeoikiZImVZ4DbwZuBGYDB+bFDgTO6U6EZmZm1g4eKmlWEtNGMDxt3qy92hCJldwU4CxJkMrwUyPiD5KuBM6U9CHgXmDfLsZo1jNc9ppZr3DDzcysh0TEXcDLakx/CNi98xGZNce/Q2hmNjIeKmlmZmad5t8hNDNrkhtuZmZm1m3+HUIzs2F4qKSZmZl1Ust/h7D4e3Tt+B3Cao389l1ZfyPPcZn1LjfcrG/5Ogszs65o+e8QFn+Prh2/Q1itkd8hLOtv5Dkus97loZLW73ydhZlZB/l3CM3MRsYNN7PBfJ2FmVmb+HcIzcxGzkMlrZ+1/DqLoikT2n+tRaPXA5Tp2gHHYtD8b2f5d7PGDP8OoVkLuSztL264WT9r+XUWRd8/5RyOmtPej1gj11lAua4dcCxm/cu/Q2hmNnIeKml9y9dZmJmZWaMkzZM0R9J1kq7K0zaWdL6kO/Lfjbodp41dwzbcJB0vabGkGwvT6iappMMlzZV0m6S3tCtws9HwdRZmZmY2Ar6pmXVNI2fcTgT2qJpWM0klbQvsB2yX3/NDSeNaFq1Z60wBLpF0PXAF8NuI+AMwC3iTpDuAN+XXZmZmZrX4pmbWMcNegBMRF0uaVjV5b2B6fn4SMAAcmqefHhH/AO6WNJc0/OxvrQnXrDV8nYWZmY1UIzeEmLn9ipW/KecbQowZbbup2bJly5i5/dPtjB1o/KZm7dIPNwVr5z6O9M4J9ZJ0c+CywnIL8rTVtOuufO1OhrIlnOMxMzMz64i23dRsYGCAoy5Z3spYa2r0pmbt0g83BWvnPrb6lneqMS1qLdiuu/K1OyHLlnCOx8zMzKz9ijc1kzTopmb5RIZvamZtNdK7Sta7894CYGphuS2AhSMPz8zMzMysu3xTMyuDkZ5xqyTpLAYn6WzgVEnfATYDtibd+MHM2qDRH96sXGvh6yzMzMxGxD8eb103bMNN0mmkG5FMlrQA+DKpwbZakkbETZLOBG4GVgCfiIj2X2lpZmZmZtYmvqmZlUEjd5Xcv86smkkaEUcCR44mKDMzMzMzM1tlpNe4mZmZmZmZWYe0+q6SZmZmLdfo9ZxFvqbTzMzGEjfczMzMzEqk2Y4Kd1KY9Qc33MzMzMzaaCRnjM3MqvkaNzMzMzMzs5Jzw83MzMzMzKzkPFTSrI/4ugkzMzOz3uQzbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcr7GzczMxqRa13TO3H4FM+pc6+lrOq1XNXP9cuUz4Hw36z0+42ZmZmZmZlZyPuNmZmZmZmZd0e4fqB9LZ5d9xs3MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5IbczcnafYCx7F0waKZmZmZma3S7pufQOfaE2Ou4WZmZjYS7vgzMxus1Y2eoX5L04bnoZJmZmZmZmYl5zNuZmZmZjYkn5EeGzoxbNDaxw03MzMzsz7jCrxZ6xQ/T40MBx1px0bfN9yaLbhO3GNimyIxK592fLFXF2julTUzMzMbXtsabpL2AL4LjAOOjYhZ7dqWWSs5d8utlY3JVl0kXabGp/O3czx0rLWcu9bLnL/WCW1puEkaBxwDvAlYAFwpaXZE3NyO7XXSnPsea7qi5y/r3jGWc9fGPuev9SrnrvUy5691SrvOuO0KzI2IuwAknQ7sDTiBreycu9bLnL8l1uw1EJ1QouH/zt0xptkz0iP5TJSoY9z5ax3Rrobb5sD8wusFwCuKC0g6CDgov1wm6bYa65kMLGlLhCP0yRHEpG+2KZikbMeo1fFs2cJ1NWLY3IXey9+R5G27VMfS5s9HU7GM1BD7ULr8bTB3W64sOeg4Bnv9N+vGUbrchSHztxTHs6gs/+NqYymuMVb2lvL/0mplzb9WamQfR5q77Wq4qca0GPQi4qfAT4dciXRVROzcysBGq2wxOZ6WGzZ3offy17HUVqZYWqQlZW87lOVYO45yxsEoy94S7cdKZYwJHFebjLrs7fH9b1g/7Gc797FdP8C9AJhaeL0FsLBN2zJrJeeu9TLnr/Uq5671MuevdUS7Gm5XAltLep6ktYD9gNlt2pZZKzl3rZc5f61XOXetlzl/rSPaMlQyIlZIOhg4j3Rb1OMj4qYRrKrjw3kaULaYHE8LtTB3oVzHwrHUVqZYRq3F+dtqZTnWjmOwUsTRgtwtxX5UKWNM4LharkVlb8/uf5P6YT/bto+KWG0IuZmZmZmZmZVIu4ZKmpmZmZmZWYu44WZmZmZmZlZypW24SdpD0m2S5ko6rI3bmSrpQkm3SLpJ0qfy9CMk3Sfpuvx4a+E9h+e4bpP0lsL0nSTNyfO+J6nW7WEbiWleXs91kq7K0zaWdL6kO/LfjToRj6RtCsfgOkmPSzqkm8en7DqVu4Xt1cvhpnOmhTGNk3StpHO7GYukDSX9StKt+fjs1s3jMtbVy8WqZaZLeqxQdnypTbGsVo5WzVcuh+ZKukHSjm2IoWb5WbVMW46HpOMlLZZ0Y2Fa3dyvem9Hy7DR6FasZSx3q+IrRRlcFZPL4zp66TNXS63ydiT/W5WonthsGdrsPklaW9IZefrlkqY1FFhElO5BurDzTuD5wFrA9cC2bdrWpsCO+fkk4HZgW+AI4DM1lt82x7M28Lwc57g87wpgN9Lvefwe2HOEMc0DJldN+xZwWH5+GPDNTsVT9X+5n/TjgF07PmV+dDJ3G8jhpnOmhTF9GjgVOHek+duiOE4CPpyfrwVs2M3jMtYf9XKxapnplbxocyyrlaNV89+ayyEBrwQub3M8K8vPThwP4LXAjsCNhWk1c79GnB0tw0Z5TLsSaxnL3ar4SlEGV8Xk8rj2cemZz9wQ+7BaeTuS/y0lqic2U4aOZJ+AjwM/zs/3A85oJK6ynnHbFZgbEXdFxD+B04G927GhiFgUEdfk50uBW4DNh3jL3sDpEfGPiLgbmAvsKmlTYP2I+Fuk/8LPgX1aGOrepEKP/HefwvROxbM7cGdE3DNMnN04PmXRsdytGCKHm8qZVsUjaQtgL+DYwuSOxyJpfVLBexxARPwzIh7tRiz9YgTlaTftDfw8ksuADXM51S6NlJ8tExEXAw9XTa6X+0UdL8NGoWuxlq3cLSpLGVwVk8vj+nrpM9eMMtRbR6zJMnQk+1Rc16+A3Rs5w1jWhtvmwPzC6wV04Ms/n6Z8OXB5nnRwHkJzfOF0aL3YNs/Pq6ePRAB/lHS1pIPytCkRsQjSFwbw7A7GU7EfcFrhdbeOT5l1JXcrqnK42ZxplaOBzwHPFKZ1I5bnAw8CJ+QhQ8dKmtilWPpOjfK0aDdJ10v6vaTt2hRCrXK0qNP/7+rys6gTxwPq535RL30OShFrScrdorKUwUUuj+sbC/ta1nprq7Vyn1a+JyJWAI8BmwwXQFkbbrVanG393QJJ6wG/Bg6JiMeBHwFbATsAi4CjhomtlTG/KiJ2BPYEPiHptUOF3oF4UPpBybcDv8yTunl8yqxr+1kjh+suWmNaS2KU9DZgcURc3ehb2hUL6XcqdwR+FBEvB5aThjZ0I5a+MkwuXkMaLvgy4PvA2W0KY7hytGP/7xrlZ1Gnjkejeulz0PVYy1DuVsVTpjK4yOVxfWNhX0tXb+2wkezTiPa3rA23BcDUwustgIXt2pikNUkF7ykR8RuAiHggIp6OiGeAn7HqFH292Bbk56OOOSIW5r+LgbPyth+oDOPJfxd3Kp5sT+CaiHggx9a141NyHc3dilo5TPM50wqvAt4uaR5puMcbJJ3cpVgWAAsionLG51ekikM3YukbdXJxpYh4PCKW5ee/A9aUNLnVcdQpR4s6+f8eVH5WxdmR45HVy/2iXvocdDXWEpW7RWUqg4tcHtfX8/ta0nprO7Ryn1a+R9J4YANWH5q5mrI23K4Etpb0vNxTuR8wux0byuNJjwNuiYjvFKYXr3V4B1C5q8xsYL98N5jnAVsDV+RTpkslvTKv8wPAOSOIZ6KkSZXnwJvztmcDB+bFDiysu63xFOxPYZhPt45PD+hY7lbUy2GazJlWxBIRh0fEFhExjbTvf46I93UplvuB+ZK2yZN2B27uRiz9YohcLC7znMo4fkm7kr6HHmpxHPXK0aLZwAeUvBJ4rDIEpg0GlZ9Vsbb9eBTUy/2ijpdho9C1WMtU7haVqQyuisvlcX299JlbTYnrre3Qyn0qrutdpM/q8GcYowR3o6n1IN3x63bSnVm+0MbtvJp0avIG4Lr8eCvwC2BOnj4b2LTwni/kuG6jcMcbYGdSst4J/ADQCOJ5PunONNcDN1X2nTTu9QLgjvx3407Ek9ezLqkisUFhWleOTy88OpW7DeRw0znT4rims+qOZl2JhTSU96p8bM4GNur2cRnLjyFy8WPAx/IyB+ey7XrgMuBf2hBHvXK0GIeAY/L/ew6wc5uOSa3ys+3Hg9RQXAQ8RerZ/VC93Ac2A35XeG9Hy7BR7mdXYi1ruVsVY9fL4Kp4XB7XPzY985mrEXvp6q0t2q+Gy9CR7BOwDmn4/FxSp8TzG4mr8mYzMzMzMzMrqbIOlTQzMzMzM7PMDTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5MZUw03SiZK+1u04zHpdKz5LkmZIumSI+QOSPjyabZiZlZmkbSRdK2mppE92aJufl3RsJ7ZlvUHSayTdVoI4XE8fpTHVcGs3SRtK+pGk+yU9IWmOpH/v0LanS1rQ4nWurDgruVjSl6qWOVDSnZLWbeW2zczq6UZltx3aUW5bz/kcMBARkyLieyNZQa1OLkl7S7pO0uOScxUqOQAAIABJREFUlki6QNI0gIj4ekS4U8xWioi/RMQ2o12PpJD0glbElMvHZyQty48Fks6UtEsr1j/MtiXps5LukPSkpHslzZK0dru3PVpuuDVI0lrAn4Atgd2ADYDPArMkfbqbsbVCRATwIeDTkrYDkPQs4NvAhyPiiVZsR9L4VqzHzMa0dlV2a31Zfz2X72btsCVwUytXmCvOPwdmkuoizwN+CDzTyu1Y7+jhutXCiFgPmAS8ErgV+Iuk3du83e8BBwEfyNveE3gDcGabtztqPd1wk/RySdfkXtkzgHXy9I0knSvpQUmP5Odb5Hn7Srq6aj0zJZ2dn79V0s15nfdJ+kxe7P3Ac4F9I+LuiHgqIv4AfBL4qqT18/vnSTo8r+MRSSdIWqewrbflXrJHJV0q6aWFefMkfUbSDZIek3RG8b1DHIe9cu/045LmSzqiMG8dSSdLeihv80pJUyQdCbwG+EHu6fhBRNwBHAkcJ2kNUmL/OiIuHCbuw/JZuaV5v99RmDdD0l8l/a+kh4GVsVl51Pss5XkfkTRX0sOSZkvaLE+flnvfxheWra4sS9L3cz7fOlRhLOmDkm7Jn5vzJG3Zjn21ntDyym5W68v6jcDpbdiW9TlJfwZez6rv2U+14rsa2AG4OyIuiGRpRPw6Iu7N6zpC0sn5+Xsk3VWoo+ypNGroWfl1zXJXyf9KWpzL7xskvaSDh8+oX6dUPpsv6VBJ9wMnSFpb0tGSFubH0cpnkFR19l/SZpJ+rVRPvluFkQ2SxikNt63U666WNFXSxXmR63MuvicvP1T9sG7doijn8YKI+BJwLPDNwjq+mz8vj+dYXpOnP0dp9NsmhWV3yvu0pqQXSLoo5++SvH0kbQ18HDggIv4WESsi4ibgncAekt6QlztR0o8lnZ/jv6hYL5H0ojzvYUm3SXp3Yd6Jko6R9Nv83sslbdXkv7+2iOjJB7AWcA/wX8CawLuAp4CvAZvkf8C6pC/nXwJn5/etDTwMvLiwrmuBd+bni4DX5OcbATvm56cDJ9WIYzywAnhLfj0PuBGYCmwM/BX4Wp63I7AYeAUwDjgwL7924b1XAJvl994CfCzPmw4sqHMspgPbkxriLwUeAPbJ8z4K/F8+FuOAnYD187wB0tm04rrGAZcDvwHuzcdvuLj3zTGvAbwHWA5smufNyMfnP/OxmtDt3PGjqc/SG4AlOQfWBr4PXJzfNw0IYHxhXStzqvC/r6z3PcBj/H/27jxerqLO+/jnCwIioBAjkYRIXCIjiKJGxMFxgoggOsZxQRARHJzIDDzomMchiAsujJkZxd1RFExAtrggecSFRa4Mys4AYREJECAkJLInOKAJv+ePqgsnne57u+/t5Zzu7/v16tftrrP07/Strq46VacOTKiz7tuBJcBLcj75BPC7Xn82fvQkP/4aWAc8BqwBPkwqox8B7gaOK6z7dOAHwP3AQ8CVwCTSCajiPr4BTM9pu9W831TgceBv8+v1ysWcjy8pvP5qjuMR4Gry70VetjkwH3gQuIk0KmNZYfnRwD3AauAWYK9ef95+dDw/F8u5mbThtxp4Qc7bXyY1DLesec/jgB8UXp+W8+WzgeXAW3N6w3IX2Cfn760B5XW26/XnOWgPGtQpc15aS2rgbJbLns8ClwHbAs8Bfgd8rpD3luXnG+X/7adIv/8vAG7nqXrsx4DFwI75f/9y4Nl5WQAvKsTXsH7ICHWL2phqjvkNpN7jLfLr9+W8+zRSL/O9wNPzsp8D/1TY9svA1/PzM4Bj8/E+HXhdTj8cuLPB5/0b4Av5+XxSWf36fDxfJf8WAFuQfgc+kON6JamutHNh2weA3fLy04Az25Inep0px5GZX08qgFRI+91whqhZd1fgwcLr/wKOz893Jv3IDjdC7iIVoM+s2ccFwLwGsdxLarmTM+zhhWX7AbcV3vdzNdvewlMVhqXA+wrL/gP49kgZvEE8XwG+nJ//Q/5cXlZnvSFqGm6FzySAWc3EXWf7awvbHgrc1ev84seI+aXhdwk4CfiPQvqWpIJ3Gs013Gr3ewVwcJ11fwEcVlhvI+BPwA69/nz86P6DzlR2R/uxPr7BdoeyfsNtpErEPOC/SRWsqaQK13BlaUfSD/3k/Hoa8MJef9Z+dC8v11k25t9q0rCyhcAfSY24+eQGHBs23LYm1W0WA98ppDcsd0mV5z/k99mo15/joD5oUKfM5eKfh8uevOw2YL/C632Apfn5zEJZ9Bpq6mXAMcD38/NbyHW4OvHUNtwa1g8ZpZ5O44bbX+X3mdIghgeBl+fn7wF+m59vnMvj3fLrU4ATge1rtv8EcFmDfZ8JfDc/n0+hsUWq/6wjle3vAf67ZtvvAJ8ubPu9mv/b79uRJ6o8VHIycE/kTyS7E0DSMyR9R9Kdkh4BLga2lrRxXm8B8F5JIg2BXBgRj+dl7yR9wHfmbtHX5vT7gO1qg1AaJjYxLx92d01Mk/PzHYA5uTv5IUkPkTLA5ML69xae/4mUUUYk6TWSLsrdww+TKigT8+JTgV8BZ+au8/+QtMlI+4vUZQxPDVUaMW5J7y90kz8EvLTw/rWfh5VPw+9SXjb8nIhYQ+rdmNLkvuvtd3Kd9XYAvlrIQw+QzvQ1+z7WpyJiKCIWR8QTEXE96Szq3+bFfyE1ol4UEesi4uqIeKTBriaSRlTUs4J0hrqZeH4QEfdHGl7zJdKZ2OGL/vcnNQAfiIi7SUMzh63L6+4kaZOIWBoRtzXzntYf2vlbHRGXRcT+EfEc0lDK15N6F+qt+xBp5NFLgS8VFjUsdyPi16Se6m8CKyWdqDzc0rquUZ3yjxHxWGHZer/XjPx7O7mmTvdx0mgFSPW7ZsumkeqHI9UtRjKF1HB7CJ68nOnmPOTxIdJ1ncPfm3NIZeoLgL2BhyPiirzsX0n5+QpJN0r6h5xetz6fbUeD+nyu/zyQj2sH4DU1x30Q8NzCti3X55tR5YbbCmBKbnwNe17+O4f0Q/qaiHgmqUCD9A8kIi4jnan4G+C9pAKTvOzKiJhF6mr+KU9dqHgB8GZJW9TE8U7SMJvLCmlTa2Janp/fTfpR37rweEZEnNHaoW/gdGARMDUingV8u3Csf4mIz0TETsBfA28lXd8B6YvRjIZx5/G+3wWOJHWlb006y1z8vzT7PtYbI32XlpMKKABy/n82abjXozm5OONosdCiwX6Xs6G7gQ/V5LHNI+J3rR+O9ZM2VnZH+7H+Y5PxjFSJmMyGlSwAImIJ8BFSb8gqSWcqXy9qA6Mjv9URcSXp8oa616BJ2pXUo3cG659MGLHcjYivRcSrSKNwXkwaQmfd16hOWZsv1vu9ZuTf2ztq/u9bRcR+heXNXo81Ur12pLrFSP4euCYiHs3Xsx1NOim2Ta5jPsxT35vHSPX0g0gdMcX6/L0R8Y8RMZk0OuNbShP7/BqYKmm34ptKmkrqYb6wkDy1sHxL0miK5fm4f1Nz3FtGxD81cXzjUuWG26Wk8b1HSXqapHeQxpJCui7rf4GHJE0APl1n+1NIZ5PWRsQlkGaOlHSQpGdFxF9I1zCsy+ufCiwDfqg0KcMmkvYhFYLHRcTDhX0fIWn7/N4fB87K6d8FDs8VEUnaQmlika2aPWili1KLD+XjfSAiHssZ8b2F9feUtEvubXyEdIZ6+JhWksY2j2akuLcgFR5/zO/3ARr8eFhpjfRdOh34gKRdlS5y/jfg8txb8EdSA+59Shcz/wMbFvbb5v1uIundpOskfl4nhm8Dx+ipGU2fldc3a1dld7Qf69/kpEdpcDJitEoEqaJSW8l6UkScHhGvI1WugsIF+DYQ2vJbLel1SpNGbZtf/xXwNtY/gTy87vB1oB8nXY8zRdI/58UNy11Jr86/+ZuQvhOPFeKx7mpUp6x1BvAJSc+RNJF0DdsP6qx3BfCI0sQmm+ff75fqqWn4vwd8TtL0XOd7mZ6aAKS23jhS/XCkusV68rZTJH0a+GA+TkjfmbWkOubTlG5ZVdvzewppSPvbiserNBnh9vnlg6Qyd11E/IGU90+TtHs+/p2BHwMXRMQFhX3vl79vmwKfI9V/7gZ+BrxY0sG5frNJ/s68pN7xtVNlG24R8WfgHaR/1oOk8aY/yYu/QrpQ8z5SQfbLOrs4ldTAOLUm/WBgqdIQy8NJ1zOQh1K+kdTKvpxUsJ4AHBsR/1mzj9OB80gXe95OulaIiLgK+EdSg/FB0kXBh7Zw2FNIDdLi44Wk2XE+K2k16YtanM70ucCPcrw3kyonwxn7q8C7lGYqajjl9khxR8RNpKEXl5K+0LuQLp61ihjpuxQRFwKfJBVoK0j57YDC5v9IOgt7P+msbG0P2eWkSSHuI00Y8a6IuL9ODGeTKrFn5u/eDaQZ/8zaUtkd5cf6d6RRFZCu0X2H0pD7F5Fuk1KMZaRKxEJSRXibXGH4P4VYd5T0hnwC5DFS+e2K8GBp12/1Q6RK6mJJa0h1nLNJ18XX+gLpOqL/yvWY9wGflzR9lHL3maRK+YOknuP7SbcHsu6rW6es4/PAVcD1pOsZr6m3bkSsA/6OPDsp6ff5e6TRA5Dqtgvzez5CutZ987zsOGCB0vDA/UepH45UTx82OefhNaTJpXYBZkbEeXn5r0jXYv6BlA8fo+bym4j4LWkyk2siYmlh0auBy/P+FwEfjog78rIj8zH/IL/3L0nXkr6zJr7TSZ0/D5CuoT4ov+dq4E2k+tBy0rDI4YliOkrrDz0dHJI2J82E88pI0+C3a79LSRcRXzDaumZmtiFJQ6TJFb4n6V2kk0MTSJXZpcDWEfE+SQeSKhLbk358zwI+GhFrla5PXkC6du3UiDhK6TYnHyNVNKaSZj37EanMfji/90TSj/VrSRWg84E3RsTrcgPxRNJMuo+SZjD757z9BZKeQWocvo30Y/59UmVhe6Upsr9H6nX+C6mxODsi6g1lMjNrW51SaYr770VEM6OsKkfp1hunR8T32rjP+aSTHp9o1z7bYZAbbh8lTYn7hjbvdyluuJmZlZ6kz5KmRH99nsDBzKw02thwO4o0U2Snb2zddXmI5/mk4fSr27jf+ZSw4VbVO62PS/4iiPSDbWZmAygiPiVpFekat3pD6s3MKk3SV0mjAA7pdSztJmkBqS7/4XY22sps1B43SSeTLvheFREvzWkTSENSppGGrewfEQ/mZceQrglYBxwVEb/qVPBmZmZmZmaDoJnJSeYD+9akzQUujIjppGkz5wJI2ol0od7OeZtv6al7p5mZmZmZmdkYjDpUMiIuljStJnkW6Y7nkC7+HiJNkTyLdJfxx4E7JC0hTf156UjvMXHixJg2bf23ePTRR9lii9pbplWTj2Xsrr766vsi3WC0tOrlX6je/93xtl/Z82+V865jbI9GMZY970K182+7DeIxQ3Xzb6O8WyWDlue6dbyj5d2xXuM2KSJWAETEiuF7iZCmqy/eR2RZTtuApNnAbIBJkybxxS+uP8vsmjVr2HLLttxkvOd8LGO355573jn6Wr01bdo0rrrqqg3Sh4aGmDlzZvcDGiPH236SSp1/q5x3HWN7NIqx7HkXqp1/220Qjxmqm38b5d0qGbQ8163jHS3vtntyEtVJq3sRXUScSJpWmRkzZkTth9FPGcLHYmZmZmZm4zHWG3CvlLQdQP67KqcvI90bZ9j2pHvZmJmZmZmZ2RiNtcdtEWla0Xn57zmF9NMlnQBMBqYDV4w3yKpbfM/DHDr33KbXXzrvLR2MxrrF/3erqlbzLjj/Wnm47DVrn2kNvktzdlnb8Hvm71TnjNpwk3QGaSKSiZKWAZ8mNdgWSjoMuAt4N0BE3ChpIXATsBY4IiLWdSh2MzMzMzOzgdDMrJIHNlhU9+7rEXE8cPx4gjIzMzMzM7OnjPUaNzMzMzMzM+sSN9zMzMzMzMxKrt23AzAzMzMzswHVaEKTRjyZSfPc42ZmZmZmZlZybriZmZWQpKmSLpJ0s6QbJX04p0+QdL6kW/PfbQrbHCNpiaRbJO3Tu+jNzMys3dxwMzMrp7XAnIh4CbA7cISknYC5wIURMR24ML8mLzsA2BnYF/iWpI17ErmZmZm1nRtuZmYlFBErIuKa/Hw1cDMwBZgFLMirLQDenp/PAs6MiMcj4g5gCbBbd6M2MzOzTvHkJGZmJSdpGvAK4HJgUkSsgNS4k7RtXm0KcFlhs2U5rXZfs4HZAJMmTWJoaGiD95u0OczZZW1LMdbbTyetWbOm6+/ZKsdoZmbt5IabmVmJSdoS+DHwkYh4RFLDVeukxQYJEScCJwLMmDEjZs6cucFGXz/tHL60uLWfh6UHbbifThoaGqJe7GXiGM3MrJ08VNLMrKQkbUJqtJ0WET/JySslbZeXbwesyunLgKmFzbcHlncrVjMzM+ssN9zMzEpIqWvtJODmiDihsGgRcEh+fghwTiH9AEmbSXo+MB24olvxmpmZWWe54WZmVk57AAcDb5B0bX7sB8wD9pZ0K7B3fk1E3AgsBG4CfgkcERHrehO6DTLfysLMrDN8jZuZWQlFxCXUv24NYK8G2xwPHN+xoMyaM3wri2skbQVcLel84FDSrSzmSZpLupXF0TW3spgMXCDpxT7xYGa2Pve4mZmZWdv4VhbWryQtlbQ4j4C4Kqc17Ek2azf3uJmZmVlHtPNWFnl/bb+dRT/cDmFQb+vQo+PeMyLuK7yeS52e5G4HZYPBDTczMzNru3bfygI6czuLbt/KohMG9bYOJTnuWcBwEAuAIdxwa8m0uee2tP7SeW/pUCTl54abmZmZtdVIt7LIvW2+lYVVUQDnSQrgO/lEQqOe5PU001tcRo16r1vt2W6nXnx2ZenVdsPNzMzM2qaJW1nMY8NbWZwu6QTS5CS+lYWV1R4RsTw3zs6X9PtmN2ymt7iMDm3QGzZnl7Ut9Wy3Uy96yUvSu+uGm5mZmbXV8K0sFku6Nqd9nNRgWyjpMOAu4N2QbmUhafhWFmvxrSyspCJief67StLZpEl0GvUkm7WdZ5W0vibpZEmrJN1QSPO9hMzMOiQiLokIRcTLImLX/Ph5RNwfEXtFxPT894HCNsdHxAsjYseI+EUv4zerR9IW+fYWSNoCeBNwA0/1JMP6PclmbeeGm/W7+cC+NWnDM0BNBy7Mr6m5l9C+wLckbdy9UM3MzKykJgGXSLqONJT33Ij4JakneW9JtwJ759dmHeGhktbXIuLiPB11UaMZoJ68lxBwh6Thewld2o1YzczMrJwi4nbg5XXS7wf26n5ENojccLNB5HsJ1VGWGZOaVbV4zczMzMbDDTezpwz0vYTKMmNSs6oWr5mZmdl4+Bo3G0Qr88xP+F5CZmZmZlYFbrjZIGo0A9Qi4ABJm0l6Pr6XkJmZmZmVhIdKWl+TdAZpIpKJkpYBn8b3EjIzMzOzinHDzfpaRBzYYFHdGaAi4njg+M5FZGZmZmbWOg+VNDMzMzMzK7lx9bhJWgqsBtYBayNihqQJwFnANGApsH9EPDi+MM3MzMzMzAZXO3rc9oyIXSNiRn49F7gwIqYDF+bXZmZmZmZmNkadGCo5C1iQny8A3t6B9zAzMzMzMxsY452cJIDzJAXwnXxT4kkRsQIgIlZI2rbehpJmA7MBJk2axNDQ0HrL16xZs0FaVU3aHObssrbp9ct83P30fzEzMzMzq4rxNtz2iIjluXF2vqTfN7thbuSdCDBjxoyYOXPmesuHhoaoTauqr592Dl9a3PxHvfSgmZ0LZpz66f9iZmZmZlYV4xoqGRHL899VwNnAbsBKSdsB5L+rxhukmZmZmZnZIBtzw03SFpK2Gn4OvAm4AVgEHJJXOwQ4Z7xBmpmZmZmZDbLx9LhNAi6RdB1wBXBuRPwSmAfsLelWYO/82szMWiDpZEmrJN1QSJsg6XxJt+a/2xSWHSNpiaRbJO3Tm6jNzMysU8Z8jVtE3A68vE76/cBe4wnKzMyYD3wDOKWQNny7lXmS5ubXR0vaCTgA2BmYDFwg6cURsa7LMZuZmVmHjHdykoEzbe65LW8zZ5fOvsfSeW9p7Q3MrPQi4mJJ02qSZwEz8/MFwBBwdE4/MyIeB+6QtIR0zfGl3YjVzMzMOs8NNzOz6mh0u5UpwGWF9ZbltA2MdisWaP0WJtD925hU4dYkgxyjpJOBtwKrIuKlOW0CcBYwDVgK7B8RD+ZlxwCHAeuAoyLiV20Pysys4txwMzOrPtVJi3orjnYrFmj9FibQ/duYVOHWJAMe43w81NfMrK3GdTsAMzPrqka3W1kGTC2stz2wvMuxmT0pIi4GHqhJnkUa4kv++/ZC+pkR8XhE3AEMD/U1Kw1JUyVdJOlmSTdK+nBOP07SPZKuzY/9eh2r9S/3uJmVxFiun/T1jQNn+HYr81j/diuLgNMlnUDqsZhOmu3XrExKOdS37MNZm1GFYbmd0OXjXgvMiYhr8u2wrpZ0fl725Yj4YrcCscHlhpuZWQlJOoM0EclEScuAT5MabAslHQbcBbwbICJulLQQuIlUuTjCw8ysQno61Lfbw3w7oQrDcjuhm8edTzoMn3hYLelmGpxgMOsUN9zMzEooIg5ssKju7VYi4njg+M5FZDZuKyVtl3vbPNTXKivP+PsK4HJgD+BISe8HriL1yj1YZ5tRe4vLqFHv9VgmsWqXXnx2ZenVdsPNzMzMusFDfa3yJG0J/Bj4SEQ8Ium/gM+Reok/B3wJ+Ifa7ZrpLe6G1i/LqN9UmLPL2pYnsWqXXvSSl6VX2w03MzMzaysP9bV+JGkTUqPttIj4CUBErCws/y7wsx6FZwPADTczMzNrKw/1tX4jScBJwM0RcUIhfbvhSXeAvwdu6EV8NhjccDMzMzMzG9kewMHAYknX5rSPAwdK2pU0VHIp8KHehGeDwA03swprday6bx9gZmbWuoi4hPozoP6827HY4PINuM3MzMzMzErODTczMzMzM7OSc8PNzMzMzMys5HyNWx/wdU5mZmZmZv3NPW5mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZXcwE9O0urEHmZmZmZm1huDPCmfe9zMzMzMzMxKzg03MzMzMzOzkhv4oZJmZmZmZlXkS34GixtuZgNkpAJ+zi5rObRmeT+NCzczMzOrMjfczMxs3Ab5YnEzM7NucMPNzMzMzMz60liGk5b15KInJzEzMzMzMys597gNoH4682Bm1eShlWZm6/NEIzaavmu4OdObmZnZIFl8z8MbTC41Gp8MMauejg2VlLSvpFskLZE0t1PvY9ZuzrtWZc6/VlXOu1Zlzr/WDR3pcZO0MfBNYG9gGXClpEURcVMr+3HvWXVVdRhUu/Juv6jq/3FQOf9aVTnv2ljqfPP33aIDkbTO+bf/1ObHerdMKupW/adTQyV3A5ZExO0Aks4EZgHOwBU1nIFHy7h9wHm3y8Z6gqbZvNhqYVrxa0D7Nv+2+iM6FiX6Pw6ivs27NhCcfwdct050d6rhNgW4u/B6GfCa4gqSZgOz88s1km6p2cdE4L4OxddVR/lYRqV/b7hoh3a/1yhGzbvQVP6Fiv3f2/G/HeH/2HbNxtuNmKqUf/sl73aiLOpAXin950jjGEuXd6Ez+beb5VYHtZzX+uG49/z36uTfJvNuZfRT3bYZ7T7esdYbOtVwU520WO9FxInAiQ13IF0VETPaHVgv+FgqZdS8C6PnX6jeZ+V4+8K4y16oxmfrGNujRDEObNnbDoN4zFCq425L2VslJfrsu6Isx9upyUmWAVMLr7cHlnfovczayXnXqsz516rKedeqzPnXuqJTDbcrgemSni9pU+AAYFGH3susnZx3rcqcf62qnHetypx/rSs6MlQyItZKOhL4FbAxcHJE3NjibvqmOxkfS2W0Ke8Oq9pn5Xgrro35twqfrWNsj1LEOOBlbzsM4jFDSY67zfm3Kkrx2XdRKY5XERsMITczMzMzM7MS6dgNuM3MzMzMzKw93HAzMzMzMzMrudI13CTtK+kWSUskze11PM2QdLKkVZJuKKRNkHS+pFvz320Ky47Jx3eLpH16E/WGJE2VdJGkmyXdKOnDOb1yx9JLVcvDkpZKWizpWklX9TqeWq1+v2x8qpB/y5hnq5BPG8R4nKR78md5raT9ehnjeFQh77Zbvf9pv2tUV7H26Zd6bbOqVP8tVcNN0sbAN4E3AzsBB0raqbdRNWU+sG9N2lzgwoiYDlyYX5OP5wBg57zNt/Jxl8FaYE5EvATYHTgix1vFY+mJCufhPSNi1zLco6SO+TT5/bLxqVj+LVuenU/58+l8NowR4Mv5s9w1In7e5ZjaomJ5t53mU/9/2s8a1VWsfebTH/XaZlWm/luqhhuwG7AkIm6PiD8DZwKzehzTqCLiYuCBmuRZwIL8fAHw9kL6mRHxeETcASwhHXfPRcSKiLgmP18N3AxMoYLH0kOVzMNl1uL3y8bH+XeMqpBPG8TYLwYy7/b5/7SuEeoq1ib9Uq9tVpXqv2VruE0B7i68XkZ1v4yTImIFpAwBbJvTK3GMkqYBrwAup+LH0mVV/EwCOE/S1ZJm9zqYJjXKkzY+Vcm/VcmzVcmnR0q6Pg+Pquqw46rkXWujmrqKddZA1AXLXv8tW8NNddL67X4FpT9GSVsCPwY+EhGPjLRqnbRSHUsPVPEz2SMiXkkaYnSEpNf3OiDrmarkX+fZ9vkv4IXArsAK4Eu9DWfMqpJ3rU1aqKtYZ/XNd68K9d+yNdyWAVMLr7cHlvcolvFaKWk7gPx3VU4v9TFK2oSUaU+LiJ/k5EoeS49U7jOJiOX57yrgbKoxxKFRnrTxqUT+rVCeLX0+jYiVEbEuIp4Avkt5P8vRVCLvWns0qKtYZ/V1XbAq9d+yNdyuBKZLer6kTUkX/i3qcUxjtQg4JD8/BDinkH6ApM0kPR+YDlzRg/g2IEnAScDNEXFCYVHljqWHKpWHJW0haavh58CbgCrMTtYoT9r4lD7/Vizu/a/nAAAgAElEQVTPlj6fDldKsr+nvJ/laEqfd609RqirWGf1bV2wUvXfiCjVA9gP+ANwG3Bsr+NpMuYzSENM/kJqhR8GPJs0A82t+e+EwvrH5uO7BXhzr+MvxPU6Ulfv9cC1+bFfFY+lx59jZfIw8ALguvy4sYzxtvr98mPcn3ep829Z82wV8mmDGE8FFudyfxGwXa8/y3EcX6nzbrf+p72OqQvHXLeu0uu4+unRL/XaFo63MvVf5Tc3MzMzMzOzkirbUEkzMzMzMzOr4YabmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuPUBSUslvbHXcZiZVY2kHSX9j6TVko7qdTxmZmaNuOFmZm0j6SBJ53Vgvz45YZ3yr8BQRGwVEV/rdTBmRZLmS/r8OLYfkvTBMW77bUmfHOt722Dw73N3ueFmZm0TEadFxJuGX0sKSS/qZUxm9Uh6Wn66A3BjL2Oxamul4lqlSm5EHB4Rn+t1HFZdhXLW2sQNtw6S9ApJ1+QhOGdJOlPS5yUdKumSmnWfrOBKekseuvOIpLslHVez7sGS7pR0v6Rju3hIZmalIOloSffk8vUWSXvV9k5ImilpWeH10rzd9cCjkn4N7Al8Q9IaSS9uovx9naTfSXooLz80p28m6YuS7pK0MvdWbN6VD8Osy1whNwBJpwLPA/5fLkP/NddnD5N0F/DrvN4PJd0r6WFJF0vaubCP+ZK+JekXeR+/lfRcSV+R9KCk30t6RWH9Dcr+rh94D7nh1iGSNgV+CpwKTAB+CLyzyc0fBd4PbA28BfgnSW/P+90J+C/gYGAy8Gxg+7YGb5UxjsrrxyRdL+lRSSdJmpQLzdWSLpC0TV53Wi6EP5ArqQ9KOlzSq/P2D0n6RmHfT56UkHRxTr4uF8bvGeVY3irp2rzP30l6WYP1NpI0V9Jt+eTFQkkT8rJfSjqyZv3rJL2jxY/WSkzSjsCRwKsjYitgH2Bpk5sfSCpXt46INwD/DRwZEVtGxB8Yufx9HvAL4OvAc4BdgWvzfv8deHFOexEwBfjU+I7Uyq5BxfVtkm7MZdmQpJc0WjenN6zUthDHrFx+PpLLxn0Li3fIleHVks6TNLGw3WgV6s/n5zMlLcu/OfcC35c0UdLP8nE+IOm/JbleOUAi4mDgLuDvImJLYGFe9LfAS0hlM6RyczqwLXANcFrNrvYHPgFMBB4HLs3rTQR+BJwA4y77+4K/YJ2zO7AJ8JWI+EtE/Ai4spkNI2IoIhZHxBMRcT1wBulLAPAu4GcRcXFEPA58EniiA/FbyY2zAHsnsDepovl3pEL146RCciOgdpKG15AK3fcAXwGOBd4I7AzsL+lva9YnIl6fn748V4rPGuFYXgmcDHyIdDLiO8AiSZvVWf0o4O2k78Rk4EHgm3nZ6aSK+fB+dyINhTu30XtbJa0DNgN2krRJRCyNiNua3PZrEXF3RPxvvYWjlL8HARdExBm5XL8/Iq6VJOAfgX+JiAciYjXwb8AB4zpKK706FdefkvLMR0iN+5+TGmqb1q4bEf+RdzNapXZEknYDTgE+Rjrh8HrW/y14L/CBvP9Ngf9bWNbKez+XdCJ6B2A2MAdYlo9zEuk3JFqJ3frWcRHx6HA5GxEnR8TqXG89Dni5pGcV1j87Iq6OiMeAs4HHIuKUiFgHnAUM97iNp+zvC264dc5k4J6IKBZidzazoaTXSLpI0h8lPQwcTqpQD+/37uF1I+JR4P42xWzVMp4C7OsRsTIi7iH1OFweEf+TC9WzeaqQHPa5iHgsIs4j9UicERGrCtvXrt+qfwS+ExGXR8S6iFhAOuu2e511PwQcGxHLCj8C71IaunM2sKukHfK6BwE/yetZn4iIJaSK8XHAKqVh6JOb3PzukRaOUv5OBep9x54DPAO4Ovc+PAT8MqfbYHkPcG5EnB8RfwG+CGwO/HWjDZqo1I7mMODk/J5PRMQ9EfH7wvLvR8QfciV6IalXeCzv/QTw6Yh4PO/rL8B2wA75RMZ/19R5bHA9Wc5K2ljSvNwT/AhPnVSYWFh/ZeH5/9Z5vSWMu+zvC264dc4KYEo+Ezvsefnvo6QfeQAkPbdm29OBRcDUiHgW8G1geD8rSJWH4W2fQeqhsAEzzgKsqUJyHOu3agdgznClN1d8p5JOVNRb9+zCejeTGrGTck/HuTzV03EALZ69tmqIiNMj4nWk/BCkoYrrla2kHoINNh1l1yOVv3cDL6yzzX2k78HOEbF1fjwr98DYYJlM4SRtRDxByjdT6q3cZKV2NI1OKAy7t/D8T+Tyegzv/cfcIzLsP4ElwHmSbpc0t4WYrX/UK1OLae8FZpFG6TwLmJbTxRg0KPsHhhtunXMpsBY4StLT8jU2u+Vl1wE7S9pV0tNJFe+irYAHIuKxPATivYVlPwLeqnSB/KbAZ/H/cWCNo/JaNncDxxcqvVtHxDMi4owG6765Zt2n594/SMOUDpT0WtKZ7ou6dAzWJUr3XntDHkr7GKnRtI50vdl+kibkE2IfGcPuRyp/TwPeKGn/XK4/W9KuuXL+XeDLkrbNMU6RtM+Gu7c+VKykLieVxwDkk7dTgXvqrAvtqdQ2OqEwmlbfe73Yc0/dnIh4AWnI/Uc1YBNFGJBO5L5ghOVbkUbQ3E+qm/zbWN9ohLJ/YLjC3yER8WfgHcChpGtw3gP8JC/7A6nBdQFwK3BJzeb/DHxW0mrSxe3DF3sSETcCR5DOCq/I+16GDZwOV17bZbQCfdh3gcPzMDVJ2kJpdr+t6qz7beD44eGQkp4jaVZh+c9JFafPAmflSrX1l82AeaSerntJ1+d8nDQZ1HWknoPzSNdGtGqk8vcuYD/StT0PkL5rL8+Ljyb1PlyWey8uAHYcw/tb9RTLuYXAW5QmitqElFceB35XZ11oT6X2JOAD+T03yicN/qqJ7cb13koTSr0oN04fIf3+DFQl2gD4AvCJPALmXXWWn0Lqhb4HuAm4bBzv1ajsHxwR4UeXHsB84PO9jsOP/ngALwOuAFaTKpE/Iw3TeTqpwvoIcD3wL8CywnZLgTcWXv+AdCHx8OsPkiZggHQGNoCnFZYvA2bWbP+J/PxQ4JLCssNJJxgeAvYf5Xj2JU3g81De5ofAVrUxk044fRS4JR/7bcC/1ezrpBz3q3v9f/LDDz/6+0Hqtborl13/F/h7UgX1YeA3pCG0jdbdEjgnl2V3kmY0DeBFef2m6g35Pa/P+1kC7JPTh4APFtZ7soxu5b2BmcXfkZz2L7lsfjT/Lnyy1/8LP/zo94cifB1pt0iaTyr4PtHrWMzMzMzMrDo8VNL6mqSTJa2SdEMhbYKk8yXdmv9uU1h2jKQlSvdE8/UpZmZmZlYKbrh1UUQc6t62rptPGoJXNBe4MCKmAxfm18P3/DqAdG+yfYFvSdq4e6H2N0kfV7rpbO3jF72OzcyszFx+mhngoZLW/yRNI920/KX59S2ka7RWSNoOGIqIHSUdAxARX8jr/Yp07delvYnczMzMzCx5Wq8DAJg4cWJMmzZtg/RHH32ULbbYovsBNaGssZU1LhhbbFdfffV9EdHum9hOiogVALnxtm1On8L6sx0to/G9d2YDswE233zzV02dOpUnnniCjTaqZie2Y++MP/zhD53Iv21ThbK3LLEMWhwdKnvbqgr5t1sG8Zih8XGXPf9WOe86xvYYc97t9ewoEcGrXvWqqOeiiy6qm14GZY2trHFFjC024KoYZ/4izYx4Q+H1QzXLH8x/vwm8r5B+EvDO0fY/nH/L/NmPxrF3RjvybycfVSh7yxLLoMVR9rwbFcm/3TKIxxzR+LjLnn+rnHcdY3uMNe+W8zS1WWetzEMkyX9X5fRlpBulDtuedDNVMzMzM7OeKsVQyUYW3/Mwh849t6Vtls57S4eisT6yCDiEdBPHQ0j3sRlOP13SCaT7oU0n3SdtTKY575o11Oz3Y84uazl07rn+fljHtFpWg8tr6wzXe200pW64mY2XpDNINw6dKGkZ8GlSg22hpMNIN0J9N0BE3ChpIenGqWuBIyJiXU8CNzMzMzMrcMPN+lpEHNhg0V4N1j8eOL5zEZmZmZmZtc7XuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmYGSJoq6SJJN0u6UdKHc/oESedLujX/3aawzTGSlki6RdI+vYve+p0bbmZmZmZmyVpgTkS8BNgdOELSTsBc4MKImA5cmF+Tlx0A7AzsC3xL0sY9idz6nhtuZmZm1laSTpa0StINhTT3WFjpRcSKiLgmP18N3AxMAWYBC/JqC4C35+ezgDMj4vGIuANYAuzW3ahtUDyt1wGYmZlZ35kPfAM4pZA23GMxT9Lc/Promh6LycAFkl4cEeu6HLPZeiRNA14BXA5MiogVkBp3krbNq00BLitstiyn1e5rNjAbYNKkSQwNDW3wfpM2hzm7rG0pxnr76aQ1a9Z0/T1b1c8xuuFmZmZmbRURF+dKb9EsYGZ+vgAYAo6m0GMB3CFpuMfi0m7EalaPpC2BHwMfiYhHJDVctU5abJAQcSJwIsCMGTNi5syZG2z09dPO4UuLW6uaLz1ow/100tDQEPViL5N+jnHU3CHpZOCtwKqIeGlOmwCcBUwDlgL7R8SDedkxwGHAOuCoiPhVy1GZmZlZvxlXjwU012vR6pnsVns4oPu9HKOpQg9DJ3TquCVtQmq0nRYRP8nJKyVtl/PudsCqnL4MmFrYfHtgeduDMqO5Hrf5eLiDmZmNYNrcc3sdglVXUz0W0FyvRatnsg8dQ97tdi/HaKrQw9AJnThupa61k4CbI+KEwqJFwCHAvPz3nEL66ZJOINV9pwNXtDUos2zUyUki4mLggZpkX6BpZmZmrViZeypwj4WV2B7AwcAbJF2bH/uRGmx7S7oV2Du/JiJuBBYCNwG/BI5wh4V1ylivcevKcIcyX6RZ1mEJZY0Lyh2bmZl1nHssrPQi4hLq9wID7NVgm+OB4zsWlFnW7slJ2jrcocwXaZZ1WEJZ44Jyx2ZmZu0j6QzSRCQTJS0DPk1qsC2UdBhwF/BuSD0WkoZ7LNbiHgszs7rG2nDzBZpmZmZWV0Qc2GCReyzMzMZorDfgHh7uABsOdzhA0maSno+HO5iZmZmZmY1bM7cD8HAHMzMzMzOzHhq14ebhDmZmZmZmZr3V7slJzMyswyQtBVYD64C1ETFD0gTgLGAasBTYPyIe7FWMZmZm1l5jvcbNzMx6a8+I2DUiZuTXc4ELI2I6cGF+bWZmZn3CPW5mZv1hFul6ZIAFwBBwdK+Cabdpc89teZul897SgUjMzMx6ww03M7PqCeA8SQF8J98Xc1JErADIt2rZtt6GkmYDswEmTZrE0NDQBuusWbOmbvpI5uyytqX1mzVp87Hvu9VjGMlYPpNOKEscZmbWfW64mZlVzx4RsTw3zs6X9PtmN8yNvBMBZsyYETNnztxgnaGhIeqlj+TQMfSINWPOLmv50uKx/VQtPWhm2+IYy2fSCWWJY9C12gPs3l8zawdf42YDS9JSSYslXSvpqpw2QdL5km7Nf7fpdZxmtSJief67Cjgb2A1YKWk7gPx3Ve8iNDMzs3Zzw80GnSd4sEqRtIWkrYafA28CbgAWAYfk1Q4BzulNhGZmZtYJHipptr6+nuDB+sIk4GxJkMrw0yPil5KuBBZKOgy4C3h3D2M0MzOzNnPDzQZZWyd4qJ00oNUJFb5+WusdJLtMeVbL29RT5QkPqhz7WETE7cDL66TfD+zV/YjMzMysG9xws0HW1gkeaicN6NRkDUXtmnyhyhMeVDl2MzMzs2b5GjcbWJ7gwczMzMyqou963DxFrzUjT+qwUUSsLkzw8FmemuBhHp7gwczMzMxKou8abmZN8gQPZmZmZlYZbrjZQPIED2ZmZmZWJb7GzczMzMzMrOTccDMzMzMzAySdLGmVpBsKaRMknS/p1vx3m8KyYyQtkXSLpH16E7UNCjfczMzMrGskLZW0WNK1kq7KaQ0rxmZdNh/YtyZtLnBhREwHLsyvkbQTcACwc97mW5I27l6oNmjG1XBz4WtmZmZjsGdE7BoRM/LruhVjs26LiIuBB2qSZwEL8vMFwNsL6WdGxOMRcQewhHRrIbOOaMfkJHtGxH2F18OF7zxJc/Pro9vwPmZmZtafZgEz8/MFwBCuO1h5TIqIFQARsULStjl9CnBZYb1lOW0DkmYDswEmTZrE0NDQhm+yOczZZW1LgdXbTyetWbOm6+/Zqn6OsROzSrrwNTMzs0YCOE9SAN+JiBNpXDFeTzOV31YrRK1WlMei05XIKlRUO6EEx606aVFvxZzPTwSYMWNGzJw5c4N1vn7aOXxpcWtV86UHbbifThoaGqJe7GXSzzGOt+HW0cJ3LGceWjXWL3wJCou6yhoXlDs2MzPrmj0iYnmuH5wv6ffNbthM5bfVCtGhc89tet2x6nTlugoV1U7o4nGvlLRdrtduB6zK6cuAqYX1tgeWdyMgG0zjbbh1tPAdy5mHVo21MC1rIVnWuKDcsVXVtBYrHEvnvaVDkZiZNScilue/qySdTbomqFHF2KwMFgGHAPPy33MK6adLOgGYDEwHruhJhDYQxjU5SbHwBdYrfAFc+JqZmdkwSVtI2mr4OfAm4AaeqhjD+hVjs66SdAZwKbCjpGWSDiM12PaWdCuwd35NRNwILARuAn4JHBER63oTuQ2CMXdn5QJ3o4hYXSh8P0vjsxJmZmY22CYBZ0uCVAc5PSJ+KelKYGGuJN8FvLuHMdoAi4gDGyzaq8H6xwPHdy4is6eMZxyiC18zW0+rQzfBwzfNBklE3A68vE76/TSoGJuZWTLmhlu/FL6+RsjMzMzMzMquszN/mJlZ5Sy+5+GuzLRnZmZmzXPDzczM+pJHVJiZWT8Z16ySZmZmZmZm1nnucTMzM7O+4aG+Ztav3ONmZmZmZmZWcu5xMzMzY+Rr4ubssnaDXhxfE2fN8vWWZtYO7nEzMzMzMzMrOfe4mVlP+Uy0mZmZ2ejc42ZmZmZmZlZybriZmZmZmZmVnIdKtmh4WFe9C9Ub8dAuq6pWhzF2Q21Mo30X/f0zMzOzfuCGm5n1NV9DZ2ZmZv3AQyXNzMzMzMxKzj1uZgOkUe9TK0N/zSxxb66ZmXWTe9zMzMzMzMxKzj1uZmYFY5mQxT0pZmbWj1qdEAz8m9hJ7nEzMzMzMzMruY71uEnaF/gqsDHwvYiY16n3Mmsn512rMuff8nJv7sicd63KnH+tGzrScJO0MfBNYG9gGXClpEURcVMn3q/fjOfeWe7CHh/nXasy51+r6oQpzrvra/X/OH/fLToUiTXD+de6pVM9brsBSyLidgBJZwKzAGfgAVHvR2ekRmVZKg8471q1Of/2mbFcX1JRzrvjsPiehzueL0r0O11Gzr8DrlsnWxQRY9pwxJ1K7wL2jYgP5tcHA6+JiCML68wGZueXOwK31NnVROC+tgfYHmWNraxxwdhi2yEintOJYOppJu/m9Hr5t8yf/Wgce2eULv9WsOwtSyyDFkfp8m5Or1r+7ZZBPGZofNyly799lHcdY3uMKe92qsdNddLWayFGxInAiSPuRLoqIma0M7B2KWtsZY0Lyh1bwah5F+rn34ocX12OvW/0XdlbllgcR8eNuezdYEf9+xk1NIjHDKU67r4rextxjO0x1hg7NavkMmBq4fX2wPIOvZdZOznvWpU5/1pVOe9alTn/Wld0quF2JTBd0vMlbQocACzq0HuZtZPzrlWZ869VlfOuVZnzr3VFR4ZKRsRaSUcCvyJNi3pyRNw4hl2N2KXcY2WNraxxQbljA8add0t/fCNw7H2gT8vessTiODqojXkX+vQzGsUgHjOU5Lj7tOxtxDG2x5hi7MjkJGZmZmZmZtY+nRoqaWZmZmZmZm3ihpuZmZmZmVnJlaLhJmlfSbdIWiJpbp3lkvS1vPx6Sa/sUlxTJV0k6WZJN0r6cJ11Zkp6WNK1+fGpLsW2VNLi/J5X1Vne9c9M0o6Fz+FaSY9I+kjNOj35vDpptPzbK/XyiKQJks6XdGv+u01h/WPyMdwiaZ9C+qvyfpbkPFVv2uPxxnqypFWSbiiktS1WSZtJOiunXy5pWruPoYrKUPaWqZwtS7k6qGXpeJW1LG63Rt+ZkcrMfiFpY0n/I+ln+XXfHHPZ82+93+myaeb3pNckPV3SFZKuyzF+pqUdRERPH6SLOG8DXgBsClwH7FSzzn7AL0j3ydgduLxLsW0HvDI/3wr4Q53YZgI/68HnthSYOMLynnxmNf/Xe0k3Euz559Xh4xwx//Ywtg3yCPAfwNz8fC7w7/n5Tjn2zYDn52PaOC+7Anhtzku/AN7cgVhfD7wSuKETsQL/DHw7Pz8AOKvX/59eP8pS9papnC1juTooZWmbPqdSlsUdONa635lGZWY/PYCPAqcP5/1+OeYq5N96v9NlezTze9LrR/792DI/3wS4HNi92e3L0OO2G7AkIm6PiD8DZwKzataZBZwSyWXA1pK263RgEbEiIq7Jz1cDNwNTOv2+bdKTz6xgL+C2iLizi+/ZC83k3zKZBSzIzxcAby+knxkRj0fEHcASYLecZ54ZEZdGKmVOKWzTNhFxMfBAB2Mt7utHwF6d6DmsmFKUvRUrZ3tRrg5KWTpeVSuLx2yE70yjMrMvSNoeeAvwvUJyvxxz6fNvg9/pUqnC70n+/ViTX26SH03PFFmGhtsU4O7C62Vs+CE3s05H5aFVryC1jGu9Nnd5/kLSzl0KKYDzJF0taXad5b3+zA4AzmiwrBefV6f0+nMeSb08MikiVkAq4IBtc3qj45iSn9emd0M7Y31ym4hYCzwMPLtjkVdD6creEpSzZSxXB6UsHa8yl8UdU/OdaVRm9ouvAP8KPFFI65djHsj820mj/J70VB7yey2wCjg/IpqOsSP3cWtRvbPetS3PZtbpGElbAj8GPhIRj9QsvoY0hGWNpP2AnwLTuxDWHhGxXNK2wPmSfp/PhjwZdp1tuvKZKd188m3AMXUW9+rz6pSe5s1RbJBHRli30XGU8fjGEmsZj6PXSlX2lqScLVW5OmBl6XgN3He89jvTz4MIJL0VWBURV0ua2et4OmDg8m8njfJ70nMRsQ7YVdLWwNmSXhoRTV07WIYet2XA1MLr7YHlY1inIyRtQvrnnxYRP6ldHhGPDHd5RsTPgU0kTex0XBGxPP9dBZxN6mYv6tlnBrwZuCYiVtYu6NXn1UG9/JxH1CCPrBwe2pX/rsqrNzqOZfl5bXo3tDPWJ7eR9DTgWZR8yEcXlKbsLUs5W8JydZDK0vEqbVncCQ2+M43KzH6wB/A2SUtJwwjfIOkH9M8xD1T+7aTRfk/KJCIeAoaAfZvdpgwNtyuB6ZKen88uHgAsqllnEfB+JbsDDw93jXdSvgbmJODmiDihwTrPHb5WRtJupM/0/g7HtYWkrYafA28CalvqPfnMsgNpMLSnF59XhzWTf7tuhDyyCDgkr3YIcE5+vgg4QGn2xeeTztxfkfPMakm75//b+wvbdFo7Yy3u613Ar/N1cIOsFGVvWcrZkparg1SWjlcpy+JOGOE706jMrLyIOCYito+IaaT/7a8j4n30zzEPTP7tpGZ+T3pN0nNyTxuSNgfeCIw0Imp9UY4ZVvYjzfxyG3BsTjscODyemoHlm3n5YmBGl+J6Hamr+nrg2vzYrya2I4EbSTMAXQb8dRfiekF+v+vye5fpM3sGqfLwrEJaTz+vLhzzBvm3148R8sizgQuBW/PfCYVtjs3HcAuFmSOBGaQK7G3ANwB1IN4zgBXAX0hnHg9rZ6zA04EfkiYyuQJ4Qa//R2V4lKHsLUs5W7ZydRDL0jZ8ZqUrizt0nI2+Mw3LzH56UJhRtZ+Ouez5t97vdK9jqhNj3e9Gr+OqifFlwP/kGG8APtXK9sOVGjMzMzMzMyupMgyVNDMzMzMzsxG44WZmZmZmZlZybriZmZmZmZmVnBtuZmZmZmZmJeeGm5mZmZmZWcm54WZmZmZmZlZybriZmZmZmZmVnBtuPSLpeZLWSNq4iXWnSQpJT+tGbGaSvi3pk+Pcx0xJy8aw3XxJnx/Pe5uZmZn1GzfcOkTSUklvbLQ8Iu6KiC0jYl034zKrJelQSZcU0yLi8Ij4XK9iMusVSTtJWiTpYUmrJV0k6a97HZeZmZkbbj3gnjMri6rlxarFa+UyWv6R9ELgt8Bi4PnAZOBs4DxJr+18hGZmZo254dYBkk4Fngf8vzwc8l/zUMfDJN0F/Lp2+KOkIUlfkHRFPtN7jqQJDfb/LEknSVoh6R5Jn29myKX1D0lzJf2oJu2rkr42Uv7IvWu/lfRlSQ8AZwHfBl6b8+pDeb31hitKmiXpWkmPSLpN0r45/QOSbs49E7dL+tAYjuUVkq7J+zgLeHph2UxJyyQdLele4PuSNpP0FUnL8+MrkjarWf/jku7LPd8HtfwBW9dJ+pikH9ekfT3/f0fK0y+U9GtJ9+f/+WmSti7sY2nOP9cDj0p6Wn59T85zt0jaK69+HHBpRBwbEQ9ExOqI+BpwKvDveX/DZffsnP9WSJpTeL+N8vfzthzTwuGyvLDtIZLuyvEe28nP1czM+ocbbh0QEQcDdwF/FxFbAgvzor8FXgLs02DT9wP/QDrLuxb4WoP1FuTlLwJeAbwJ+GBbgreqOAPYT9IzAXIldn/gdEbPH68Bbge2Bd4HHE6qrG4ZEVtTQ9JuwCnAx4CtgdcDS/PiVcBbgWcCHwC+LOmVzR6EpE2Bn5IqxhOAHwLvrFntuXnZDsBs4Fhgd2BX4OXAbsAnatafCEwBDgFOlAgEohIAABJ1SURBVLRjszFZz/wA2He40ZVPar2HlDdGytMCvkAqN18CTCU1wIoOBN5Cyr8vBI4EXh0RW5HK46V5vb1JebDWQmAPSc8opO0JTM+xzNVTQ+OPAt5OKu8nAw8C36zZ3+uAHYG9gE9JeknDT8XMzCxzw627jouIRyPifxssPzUiboiIR4FPAvvX9qRJmgS8GfhI3tcq4MvAAR2N3EolIu4EriFVEAHeAPwJuIPR88fyiPh6RKwdIS8WHQacHBHnR8QTEXFPRPw+x3FuRNwWyW+A84C/aeFQdgc2Ab4SEX+JiB8BV9as8wTw6Yh4PMd7EPDZiFgVEX8EPgMcXLPNJ/P6vwHOJTVqrcQiYgVwMfDunLQvcB+wjBHydEQsyXnz8ZwfTiA1moq+FhF35/yzDtgM2EnSJhGxNCJuy+tNBFbUCW8F6fdym0LaZ3I8i4HvkxqHAB8Cjo2IZRHxOKkR+S6tP0zzMxHxvxFxHXAd6QSEmZnZiHy9SHfd3cLyO0kV2ok16+yQ01dIGk7bqIl9W/85nVRZPAV4b37dTP5oNa9MBX5eb4GkNwOfBl6c3+cZpOuDmjUZuCciopB2Z806f4yIx2q2ubNm/cmF1w/mkx+Nllt5LQD+CfguqTf4VEbJ05K2JY1O+Btgq7zswZr9PpnnI2KJpI+QGlQ7S/oV8NGIWE5qKG5XJ67tSCcQHiT1VK+3T1Ie2yU/3wE4W9ITheXrgEmF1/cWnv8J2LLOe5qZma3HPW6dE02mFU0tPH8e8BdSRaLobuBxYGJEbJ0fz4yIncceqlXUD4GZkrYH/p7UcGsmf9Tmw9Hy5d2k4WXrydeV/Rj4IjApD7P8OWnoWrNWAFNUqJGT8v5I8S0nVY6L6y8vvN5G0hYjLLfy+inwMkkvJQ3BPY3R8/QXSHnkZRHxTFKDrzYPrpeHIuL0iHgdKR8F+fo14AKe6vEr2p80nPhPhbTa8no4j90NvLkQ69YR8fSIuKfZD8HMzKweN9w6ZyXwgha3eZ/SVNTPAD4L/Kj2dgF5ONF5wJckPTNfCP9CSbVDg6zP5WFhQ6RhWndExM1jzB8rge3z9Wb1nAR8QNJeeX9TJP0VsClpyNkfgbW59+1NLR7GpaRrl47Kk0a8g3TN2kjOAD4h6TmSJgKfIl0fVfQZSZtK+htSA6DedUtWMrln9UekkxBX5NumjJantwLWAA9JmkK6FrMhSTtKekM+8fAYMDx8EtKw27+WdLykCZK2kvR/SNcfH12zq09KeoaknUnXd56V078NHC9ph/x+z5E0a8wfipmZWeaGW+d8gVS5fAh4V5PbnArMJw2jeTrpIvd63k+qNN9EGrrzI+oP77H+dzrwxvx3WKv549fAjcC9kmp7eImIK8gTjwAPA78BdoiI1aQ8ujC/z3uBRa0EHxF/Bt4BHJr38R7gJ6Ns9nngKuB60rDMa3LasHvzvpaTemwOH74mzyphAWnY4amFtJHy9GeAV5Ly5rmMnn82A+aRRjPcSxr6+HGAiLiVNHHIy0kTlqwgTZazT0T8tmY/vwGWABcCX4yI83L6V0nfg/MkrQYuI00IZGZmNi5a/9IS6xVJQ8APIuJ7vY7FrKokzSR9j7bvdSw2NpKeB/weeG5EPNLreGpJmkaaBGiTiFjb22jMzGyQuMfNzMxKQdJGwEeBM8vYaDMzM+slzyppZh2Ve1BuarB4p4i4q5vxWDnlCWVWkmZo3LfH4ZiZmZWOh0qamZmZmZmVnIdKmpmVkKSTJa2SdEMhbYKk8yXdmv9uU1h2jKQlkm6RtE9vojaz/9/e/cfIcZYHHP8+mB+iLhUgk2sUp1z+cBGhEbScTFCk6tIIMKTC/EGQowJOleooSiSQIpFL/wAJCcmq1KpqRX9YJcIRkNQSpLEafoXQE/xBi+00bRKCVTeY9GorVgAlcUCgg6d/7FyyWe+eb/d2dt6d/X6k0+3OzN4+78xzo3123nlfSapLEVfcduzYkfPz8+ctf/bZZ9m+ffv5LyhAqbGVGheMFtvx48efzMzX1BTSWPTmb8nHYFza3sZxtW8r+RsRv09nmPs7MvN3qmV/Dvw4Mw9ExDLwqsy8NSIupzNNwm46k41/A/jt3ulEeg0697Zd2/N3kGHaPQ3nXkmaNUXc4zY/P8+xY8fOW76yssLi4uLkA9qEUmMrNS4YLbaI+GE90YxPb/6WfAzGpe1tHFf7tpK/mfmtagTDbnuBxerxITrz+N1aLb8rM38O/CAiTtIp4r6z0XsMOve2Xdvzd5Bh2j0N515JmjVFFG6SpE2ZqyakJjPPRMRF1fJL6MwXtm61WnaeiFgClgDm5uZYWVmpL9pCnTt3znZLkqaOhZskTb/os6xvP/jMPAgcBFhYWEivPM2OWW23JLVF0YXbQ//3FDcs3zvUa04duLamaKThzJu7Gr8nIuLi6mrbxcDZavkqcGnXdjuB0xOPbsr5PytJKpmjSkrS9DgC7K8e7wfu6Vq+LyJeFhGXAbuA7zYQnyRJqknRV9wkaVZFxJ10BiLZERGrwCeAA8DhiLgReBy4DiAzH4mIw3QmOl8DbrrQiJKSJGm6WLhJUoEy8/oBq64ZsP2ngE/VF5EkSWqSXSUlSZIkqXAWbpIkSZJUuNZ1lXRUMEmSJElt4xU3SZIkSSqchZskSZIkFa51XSWlaTVsN1+wq68kSdKs8IqbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwFm6SJEmSVDgLN0mSJEkqnIWbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4V7cdABSUyLiFPAM8EtgLTMXIuLVwD8B88Ap4H2Z+ZOmYpQkSZLAK27S1Zn5psxcqJ4vA/dn5i7g/uq5JEmS1CgLN+mF9gKHqseHgPc0GIskSZIE2FVSsy2Br0dEAv+QmQeBucw8A5CZZyLion4vjIglYAlgbm6OlZWV59adO3eOlZUVbrlire74X/C+k7TexrZqe/skSdL0sXDTLLsqM09Xxdl9EfH9zb6wKvIOAiwsLOTi4uJz61ZWVlhcXOSG5XvHHe95Tv3R4gW3qcN6G9uq7e2TJEnTx66SmlmZebr6fRa4G9gNPBERFwNUv882F6EkSZLUYeGmmRQR2yPiFeuPgbcDDwNHgP3VZvuBe5qJUJIkSXqeXSU1q+aAuyMCOv8HX8jMr0bEUeBwRNwIPA5c12CMkiRJErDFws15sDStMvMx4I19lv8IuGbyEUmSJEmDjaOrpPNgSZIkSVKN6ugquRdYrB4fAlaAW2t4H2nmzQ85cuWpA9fWFIkmyd4OkiTNnq0WbrXMg7Vu7uXUPhfWqHM1lTrPU6lxQdmxSVPo6sx8suv5em+HAxGxXD33SzNJklpiq4VbLfNgrfubz9/DXzxU7/gpo86DVeo8T6XGBWXHJrWAvR0kSWqxLVVF3fNgRcQL5sGqrrY5D5YkjV+tvR3abtDV/2F7eEzbvrPXgyRNt5ELt2ruqxdl5jNd82B9kufnwTrAFMyD5T1CmiXme2vU2tuh7QZd/b9h2P+PEXtsNMVeD5I03bZyxc15sCSpAfZ22JxBX1TccsXa0EWaJElNG7lwcx4sSZq8tvR2kCRJw6l35A9J0rjZ20GSpBlk4SZJU8TeDtNr2HtMwftMJUnPe1HTAUiSJEmSNmbhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwjio5pPVRwYaZwNVRwSSpfUYZJVKSpFF5xU2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwFm6SJEmSVDinA5A00KDhzgdNh+HUF6qLQ+9LkmadV9wkSZIkqXAWbpIkSZJUOAs3SZIkSSqc97hJGptR7kPyvjhJkqQL84qbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwTgdQoGGHVHc4dU0z870dPI6SJNXLK26SJEmSVDivuEmSJm6UydolSZplXnGTJEmSpMJ5xa0Fur+5vuWKNW6o4Zts70eRJEmSmmPhJmmqOAhG/ezGKElSeWrrKhkReyLiREScjIjlut5HGjdzV9PM/JUkqZ1queIWEduATwNvA1aBoxFxJDO/V8f7qX6zcpXD3NX88r1DdzkuJd/N3/YZ59XPfnldSu5Kki6srq6Su4GTmfkYQETcBewF/PAwI/p92Njow3BBHx7MXU0z81eSpJaKzBz/H414L7AnM/+kev4B4C2ZeXPXNkvAUvX0dcCJPn9qB/Dk2AMcj1JjKzUuGC2212bma+oIpp/N5G61fKP8LfkYjEvb2ziu9hWXv5s897Zd2/N3kGHaPdHclSRdWF1X3KLPshdUiJl5EDi44R+JOJaZC+MMbFxKja3UuKDs2LpcMHdh4/ydknZuSdvbOMXtG8u5t+2m+Phuyay2W5Laoq7BSVaBS7ue7wRO1/Re0jiZu5pm5q8kSS1VV+F2FNgVEZdFxEuBfcCRmt5LGidzV9PM/JUkqaVq6SqZmWsRcTPwNWAbcHtmPjLCnyq5O0+psZUaF5QdGzC23C2+nWPQ9jZOZfvGeO5tu6k8vmMwq+2WpFaoZXASSZIkSdL41DYBtyRJkiRpPCzcJEmSJKlwRRRuEbEnIk5ExMmIWO6zPiLir6v1/xURvzehuC6NiH+NiEcj4pGI+EifbRYj4qmIeLD6+fiEYjsVEQ9V73msz/qJ77OIeF3XfngwIp6OiI/2bNPI/pqUC+VyqQblekS8OiLui4j/rn6/qus1t1XtPBER7+ha/uYqN09WOdhviPqJi4htEfEfEfEv1fPWtE2D9TtXbnTsp1lE3B4RZyPi4a5lQ+e5JKlMjRduEbEN+DTwTuBy4PqIuLxns3cCu6qfJeDvJhTeGnBLZr4euBK4qU9sAN/OzDdVP5+cUGwAV1fv2W9enonvs8w8sb4fgDcDPwXu7rNpU/urVpvM5VINyvVl4P7M3AXcXz2nWrcPeAOwB/jbqv3QybUlns+/PZNsyAY+Ajza9bxNbdPGes+VfY99C3yW83NylDyXJBWo8cIN2A2czMzHMvMXwF3A3p5t9gJ3ZMe/Aa+MiIvrDiwzz2TmA9XjZ+h86Luk7vcdk0b2WZdrgP/JzB9O8D2btplcLtIGub4XOFRtdgh4T/V4L3BXZv48M38AnAR2Vzn2G5n5neyMfHRH12saExE7gWuBf+xa3Iq2aSSDjv1Uy8xvAT/uWTxUnk8kUEnSSEoo3C4B/rfr+SrnF0eb2aZWETEP/C7w731WvzUi/jMivhIRb5hQSAl8PSKOR8RSn/VN77N9wJ0D1jWxvyah6X0+Fj25PpeZZ6BT3AEXVZsNausl1ePe5U37K+BjwK+6lrWlbdpYv3PloGPfRsPmuSSpULXM4zakfveI9M5RsJltahMRvw58EfhoZj7ds/oB4LWZeS4i3gX8M50uVHW7KjNPR8RFwH0R8f3q29bnwu7zmonss+hM/Ptu4LY+q5vaX5PQaJ6OQ2+ub3AL16C2FrcPIuIPgbOZeTwiFjfzkj7LimybNuW8c2XTARXCfJakKVPCFbdV4NKu5zuB0yNsU4uIeAmdD7Kfz8wv9a7PzKcz81z1+MvASyJiR91xZebp6vdZOveR9XZxaWyf0bnH64HMfKJ3RVP7a0Ka3OdbNiDXn1jvYlv9PlstH9TW1epx7/ImXQW8OyJO0em++gcR8Tna0TZdwIBz5aBj30bD5rkkqVAlFG5HgV0RcVl1pWYfcKRnmyPAB6PjSuCp9a4fdapGjPsM8Ghm/uWAbX5zfWS5iNhNZ5/+qOa4tkfEK9YfA28HHu7ZrJF9VrmeAd0km9hfE7SZXC7SBrl+BNhfPd4P3NO1fF9EvCwiLqNz1fS7VY49ExFXVn/zg12vaURm3paZOzNzns4x+WZmvp8WtE0b2+BcOejYt9FQed5AfJKkTWq8q2RmrkXEzcDXgG3A7Zn5SET8abX+74EvA++ic/P0T4E/nlB4VwEfAB6KiAerZX8G/FZXbO8FPhwRa8DPgH3VwAV1mgPuruqfFwNfyMyvlrDPIuLXgLcBH+pa1h1XE/trIgblcsNhbdagXD8AHI6IG4HHgesAqv/Rw8D36IxIeVNm/rJ63YfpjG73cuAr1U+J2tw2dQw6Vx6lz7GfdhFxJ7AI7IiIVeATjJbnkqQCRUs+M0uSJElSa5XQVVKSJEmStAELN0mSJEkqnIWbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4f4fTcC9QhRPDaEAAAAASUVORK5CYII=\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "ski_data.hist(figsize=(15, 10))\n", + "plt.subplots_adjust(hspace=0.5);" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "G00BFm1UTNrs" + }, + "source": [ + "These distributions are much better. There are clearly some skewed distributions, so keep an eye on `fastQuads`, `fastSixes`, and perhaps `trams`. These lack much variance away from 0 and may have a small number of relatively extreme values. Models failing to rate a feature as important when domain knowledge tells you it should be is an issue to look out for, as is a model being overly influenced by some extreme values. If you build a good machine learning pipeline, hopefully it will be robust to such issues, but you may also wish to consider nonlinear transformations of features." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "wgkJdG-FTNrt" + }, + "source": [ + "## 2.10 Population data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "wvMtXIAITNrt" + }, + "source": [ + "Population and area data for the US states can be obtained from [wikipedia](https://simple.wikipedia.org/wiki/List_of_U.S._states). Listen, you should have a healthy concern about using data you \"found on the Internet\". Make sure it comes from a reputable source. This table of data is useful because it allows you to easily pull and incorporate an external data set. It also allows you to proceed with an analysis that includes state sizes and populations for your 'first cut' model. Be explicit about your source (we documented it here in this workflow) and ensure it is open to inspection. All steps are subject to review, and it may be that a client has a specific source of data they trust that you should use to rerun the analysis." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qcblDsC8TNrt" + }, + "outputs": [], + "source": [ + "#Code task 29#\n", + "#Use pandas' `read_html` method to read the table from the URL below\n", + "states_url = 'https://simple.wikipedia.org/w/index.php?title=List_of_U.S._states&oldid=7168473'\n", + "usa_states = pd.___(___)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "itwPNvegTNrt", + "outputId": "51acb8ba-9199-402a-87fe-d755f941ace6" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "list" + ] + }, + "execution_count": 45, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "type(usa_states)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "hNtA2XRQTNrt", + "outputId": "50b91464-55f4-4aa4-d443-66689b340b07" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "1" + ] + }, + "execution_count": 46, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "len(usa_states)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "hjDf9_NwTNru", + "outputId": "23529399-33e8-47a9-9de0-a68b97125d75" + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Name &postal abbs. [1]CitiesEstablished[upper-alpha 1]Population[upper-alpha 2][3]Total area[4]Land area[4]Water area[4]Numberof Reps.
Name &postal abbs. [1]Name &postal abbs. [1].1CapitalLargest[5]Established[upper-alpha 1]Population[upper-alpha 2][3]mi2km2mi2km2mi2km2Numberof Reps.
0AlabamaALMontgomeryBirminghamDec 14, 181949031855242013576750645131171177545977
1AlaskaAKJuneauAnchorageJan 3, 195973154566538417233375706411477953947432453841
2ArizonaAZPhoenixPhoenixFeb 14, 1912727871711399029523411359429420739610269
3ArkansasARLittle RockLittle RockJun 15, 183630178045317913773252035134771114329614
4CaliforniaCASacramentoLos AngelesSep 9, 18503951222316369542396715577940346679162050153
\n", + "
" + ], + "text/plain": [ + " Name &postal abbs. [1] Cities \\\n", + " Name &postal abbs. [1] Name &postal abbs. [1].1 Capital Largest[5] \n", + "0 Alabama AL Montgomery Birmingham \n", + "1 Alaska AK Juneau Anchorage \n", + "2 Arizona AZ Phoenix Phoenix \n", + "3 Arkansas AR Little Rock Little Rock \n", + "4 California CA Sacramento Los Angeles \n", + "\n", + " Established[upper-alpha 1] Population[upper-alpha 2][3] Total area[4] \\\n", + " Established[upper-alpha 1] Population[upper-alpha 2][3] mi2 \n", + "0 Dec 14, 1819 4903185 52420 \n", + "1 Jan 3, 1959 731545 665384 \n", + "2 Feb 14, 1912 7278717 113990 \n", + "3 Jun 15, 1836 3017804 53179 \n", + "4 Sep 9, 1850 39512223 163695 \n", + "\n", + " Land area[4] Water area[4] Numberof Reps. \n", + " km2 mi2 km2 mi2 km2 Numberof Reps. \n", + "0 135767 50645 131171 1775 4597 7 \n", + "1 1723337 570641 1477953 94743 245384 1 \n", + "2 295234 113594 294207 396 1026 9 \n", + "3 137732 52035 134771 1143 2961 4 \n", + "4 423967 155779 403466 7916 20501 53 " + ] + }, + "execution_count": 47, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "usa_states = usa_states[0]\n", + "usa_states.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Db_ica1wTNru" + }, + "source": [ + "Note, in even the last year, the capability of `pd.read_html()` has improved. The merged cells you see in the web table are now handled much more conveniently, with 'Phoenix' now being duplicated so the subsequent columns remain aligned. But check this anyway. If you extract the established date column, you should just get dates. Recall previously you used the `.loc` accessor, because you were using labels. Now you want to refer to a column by its index position and so use `.iloc`. For a discussion on the difference use cases of `.loc` and `.iloc` refer to the [pandas documentation](https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3AffA8DpTNru" + }, + "outputs": [], + "source": [ + "#Code task 30#\n", + "#Use the iloc accessor to get the pandas Series for column number 4 from `usa_states`\n", + "#It should be a column of dates\n", + "established = usa_sates.___[:, 4]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "w3489xu3TNru", + "outputId": "4bad1d20-b93b-4b55-ace8-8703106ad1b1" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "0 Dec 14, 1819\n", + "1 Jan 3, 1959\n", + "2 Feb 14, 1912\n", + "3 Jun 15, 1836\n", + "4 Sep 9, 1850\n", + "5 Aug 1, 1876\n", + "6 Jan 9, 1788\n", + "7 Dec 7, 1787\n", + "8 Mar 3, 1845\n", + "9 Jan 2, 1788\n", + "10 Aug 21, 1959\n", + "11 Jul 3, 1890\n", + "12 Dec 3, 1818\n", + "13 Dec 11, 1816\n", + "14 Dec 28, 1846\n", + "15 Jan 29, 1861\n", + "16 Jun 1, 1792\n", + "17 Apr 30, 1812\n", + "18 Mar 15, 1820\n", + "19 Apr 28, 1788\n", + "20 Feb 6, 1788\n", + "21 Jan 26, 1837\n", + "22 May 11, 1858\n", + "23 Dec 10, 1817\n", + "24 Aug 10, 1821\n", + "25 Nov 8, 1889\n", + "26 Mar 1, 1867\n", + "27 Oct 31, 1864\n", + "28 Jun 21, 1788\n", + "29 Dec 18, 1787\n", + "30 Jan 6, 1912\n", + "31 Jul 26, 1788\n", + "32 Nov 21, 1789\n", + "33 Nov 2, 1889\n", + "34 Mar 1, 1803\n", + "35 Nov 16, 1907\n", + "36 Feb 14, 1859\n", + "37 Dec 12, 1787\n", + "38 May 29, 1790\n", + "39 May 23, 1788\n", + "40 Nov 2, 1889\n", + "41 Jun 1, 1796\n", + "42 Dec 29, 1845\n", + "43 Jan 4, 1896\n", + "44 Mar 4, 1791\n", + "45 Jun 25, 1788\n", + "46 Nov 11, 1889\n", + "47 Jun 20, 1863\n", + "48 May 29, 1848\n", + "49 Jul 10, 1890\n", + "Name: (Established[upper-alpha 1], Established[upper-alpha 1]), dtype: object" + ] + }, + "execution_count": 49, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "established" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "XAF6ZiG-TNrv" + }, + "source": [ + "Extract the state name, population, and total area (square miles) columns." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "eVYiJLUYTNrv" + }, + "outputs": [], + "source": [ + "#Code task 31#\n", + "#Now use the iloc accessor again to extract columns 0, 5, and 6 and the dataframe's `copy()` method\n", + "#Set the names of these extracted columns to 'state', 'state_population', and 'state_area_sq_miles',\n", + "#respectively.\n", + "usa_states_sub = usa_states.___[:, [___]].copy()\n", + "usa_states_sub.columns = [___]\n", + "usa_states_sub.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "WSTrL07ITNrv" + }, + "source": [ + "Do you have all the ski data states accounted for?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "RmM-xaKoTNrv" + }, + "outputs": [], + "source": [ + "#Code task 32#\n", + "#Find the states in `state_summary` that are not in `usa_states_sub`\n", + "#Hint: set(list1) - set(list2) is an easy way to get items in list1 that are not in list2\n", + "missing_states = ___(state_summary.state) - ___(usa_states_sub.state)\n", + "missing_states" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "521nVhMtTNrv" + }, + "source": [ + "No??" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qArNVHaRTNrw" + }, + "source": [ + "If you look at the table on the web, you can perhaps start to guess what the problem is. You can confirm your suspicion by pulling out state names that _contain_ 'Massachusetts', 'Pennsylvania', or 'Virginia' from usa_states_sub:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "h6Jy3tQvTNrw", + "outputId": "38ffc16c-6d22-4304-c08b-82c93b74488d" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "20 Massachusetts[upper-alpha 3]\n", + "37 Pennsylvania[upper-alpha 3]\n", + "38 Rhode Island[upper-alpha 4]\n", + "45 Virginia[upper-alpha 3]\n", + "47 West Virginia\n", + "Name: state, dtype: object" + ] + }, + "execution_count": 52, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "usa_states_sub.state[usa_states_sub.state.str.contains('Massachusetts|Pennsylvania|Rhode Island|Virginia')]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iXFAZmyPTNrw" + }, + "source": [ + "Delete square brackets and their contents and try again:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "inCZDuhTTNrw" + }, + "outputs": [], + "source": [ + "#Code task 33#\n", + "#Use pandas' Series' `replace()` method to replace anything within square brackets (including the brackets)\n", + "#with the empty string. Do this inplace, so you need to specify the arguments:\n", + "#to_replace='\\[.*\\]' #literal square bracket followed by anything or nothing followed by literal closing bracket\n", + "#value='' #empty string as replacement\n", + "#regex=True #we used a regex in our `to_replace` argument\n", + "#inplace=True #Do this \"in place\"\n", + "usa_states_sub.state.___(to_replace=___, value=__, regex=___, inplace=___)\n", + "usa_states_sub.state[usa_states_sub.state.str.contains('Massachusetts|Pennsylvania|Rhode Island|Virginia')]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "WNh71rW1TNrx" + }, + "outputs": [], + "source": [ + "#Code task 34#\n", + "#And now verify none of our states are missing by checking that there are no states in\n", + "#state_summary that are not in usa_states_sub (as earlier using `set()`)\n", + "missing_states = ___(state_summary.state) - ___(usa_states_sub.state)\n", + "missing_states" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "br2Ng2gxTNrx" + }, + "source": [ + "Better! You have an empty set for missing states now. You can confidently add the population and state area columns to the ski resort data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "LQDbrIjgTNrx" + }, + "outputs": [], + "source": [ + "#Code task 35#\n", + "#Use 'state_summary's `merge()` method to combine our new data in 'usa_states_sub'\n", + "#specify the arguments how='left' and on='state'\n", + "state_summary = state_summary.___(usa_states_sub, ___=___, ___=___)\n", + "state_summary.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "CG46I0-ZTNrx" + }, + "source": [ + "Having created this data frame of summary statistics for various states, it would seem obvious to join this with the ski resort data to augment it with this additional data. You will do this, but not now. In the next notebook you will be exploring the data, including the relationships between the states. For that you want a separate row for each state, as you have here, and joining the data this soon means you'd need to separate and eliminate redundances in the state data when you wanted it." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "p4NAsVeiTNrx" + }, + "source": [ + "## 2.11 Target Feature" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Z9PKLXuITNrx" + }, + "source": [ + "Finally, what will your target be when modelling ticket price? What relationship is there between weekday and weekend prices?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "B4_3RAjtTNry" + }, + "outputs": [], + "source": [ + "#Code task 36#\n", + "#Use ski_data's `plot()` method to create a scatterplot (kind='scatter') with 'AdultWeekday' on the x-axis and\n", + "#'AdultWeekend' on the y-axis\n", + "ski_data.___(x=___, y=___, kind=___);" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "tc5AkUM9TNry" + }, + "source": [ + "A couple of observations can be made. Firstly, there is a clear line where weekend and weekday prices are equal. Weekend prices being higher than weekday prices seem restricted to sub $100 resorts. Recall from the boxplot earlier that the distribution for weekday and weekend prices in Montana seemed equal. Is this confirmed in the actual data for each resort? Big Mountain resort is in Montana, so the relationship between these quantities in this state are particularly relevant." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "neqpTpj4TNry" + }, + "outputs": [], + "source": [ + "#Code task 37#\n", + "#Use the loc accessor on ski_data to print the 'AdultWeekend' and 'AdultWeekday' columns for Montana only\n", + "ski_data.___[ski_data.state == ___, [___, ___]]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "m0fGJ8POTNry" + }, + "source": [ + "Is there any reason to prefer weekend or weekday prices? Which is missing the least?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "oAKRmA-HTNry", + "outputId": "c1f10496-0d26-4bcb-e02e-0e84bfb982da" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "AdultWeekend 4\n", + "AdultWeekday 7\n", + "dtype: int64" + ] + }, + "execution_count": 58, + "metadata": {}, + "output_type": "execute_result" + } ], - "text/plain": [ - " Name Region state summit_elev vertical_drop \\\n", - "104 Crystal Mountain Michigan Michigan 1132 375 \n", - "295 Crystal Mountain Washington Washington 7012 3100 \n", - "\n", - " base_elev trams fastEight fastSixes fastQuads ... LongestRun_mi \\\n", - "104 757 0 0.0 0 1 ... 0.3 \n", - "295 4400 1 NaN 2 2 ... 2.5 \n", - "\n", - " SkiableTerrain_ac Snow Making_ac daysOpenLastYear yearsOpen \\\n", - "104 102.0 96.0 120.0 63.0 \n", - "295 2600.0 10.0 NaN 57.0 \n", - "\n", - " averageSnowfall AdultWeekday AdultWeekend projectedDaysOpen \\\n", - "104 132.0 54.0 64.0 135.0 \n", - "295 486.0 99.0 99.0 NaN \n", - "\n", - " NightSkiing_ac \n", - "104 56.0 \n", - "295 NaN \n", - "\n", - "[2 rows x 27 columns]" - ] - }, - "execution_count": 11, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data[ski_data['Name'] == 'Crystal Mountain']" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "So there are two Crystal Mountain resorts, but they are clearly two different resorts in two different states. This is a powerful signal that you have unique records on each row." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.3.2 Region And State" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What's the relationship between region and state?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You know they are the same in many cases (e.g. both the Region and the state are given as 'Michigan'). In how many cases do they differ?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 10#\n", - "#Calculate the number of times Region does not equal state\n", - "(ski_data.Region ___ ski_data.state).___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You know what a state is. What is a region? You can tabulate the distinct values along with their respective frequencies using `value_counts()`." - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "New York 33\n", - "Michigan 29\n", - "Sierra Nevada 22\n", - "Colorado 22\n", - "Pennsylvania 19\n", - "Wisconsin 16\n", - "New Hampshire 16\n", - "Vermont 15\n", - "Minnesota 14\n", - "Montana 12\n", - "Idaho 12\n", - "Massachusetts 11\n", - "Washington 10\n", - "Maine 9\n", - "New Mexico 9\n", - "Wyoming 8\n", - "Utah 7\n", - "Oregon 6\n", - "Salt Lake City 6\n", - "North Carolina 6\n", - "Connecticut 5\n", - "Ohio 5\n", - "West Virginia 4\n", - "Virginia 4\n", - "Mt. Hood 4\n", - "Illinois 4\n", - "Alaska 3\n", - "Iowa 3\n", - "Missouri 2\n", - "Arizona 2\n", - "Indiana 2\n", - "South Dakota 2\n", - "New Jersey 2\n", - "Nevada 2\n", - "Rhode Island 1\n", - "Maryland 1\n", - "Tennessee 1\n", - "Northern California 1\n", - "Name: Region, dtype: int64" - ] - }, - "execution_count": 13, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data['Region'].value_counts()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "A casual inspection by eye reveals some non-state names such as Sierra Nevada, Salt Lake City, and Northern California. Tabulate the differences between Region and state. On a note regarding scaling to larger data sets, you might wonder how you could spot such cases when presented with millions of rows. This is an interesting point. Imagine you have access to a database with a Region and state column in a table and there are millions of rows. You wouldn't eyeball all the rows looking for differences! Bear in mind that our first interest lies in establishing the answer to the question \"Are they always the same?\" One approach might be to ask the database to return records where they differ, but limit the output to 10 rows. If there were differences, you'd only get up to 10 results, and so you wouldn't know whether you'd located all differences, but you'd know that there were 'a nonzero number' of differences. If you got an empty result set back, then you would know that the two columns always had the same value. At the risk of digressing, some values in one column only might be NULL (missing) and different databases treat NULL differently, so be aware that on many an occasion a seamingly 'simple' question gets very interesting to answer very quickly!" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 11#\n", - "#Filter the ski_data dataframe for rows where 'Region' and 'state' are different,\n", - "#group that by 'state' and perform `value_counts` on the 'Region'\n", - "(ski_data[ski_data.___ ___ ski_data.___]\n", - " .groupby(___)[___]\n", - " .value_counts())" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The vast majority of the differences are in California, with most Regions being called Sierra Nevada and just one referred to as Northern California." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.3.3 Number of distinct regions and states" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 12#\n", - "#Select the 'Region' and 'state' columns from ski_data and use the `nunique` method to calculate\n", - "#the number of unique values in each\n", - "ski_data[[___, ___]].___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Because a few states are split across multiple named regions, there are slightly more unique regions than states." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.3.4 Distribution Of Resorts By Region And State" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If this is your first time using [matplotlib](https://matplotlib.org/3.2.2/index.html)'s [subplots](https://matplotlib.org/3.2.2/api/_as_gen/matplotlib.pyplot.subplots.html), you may find the online documentation useful." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 13#\n", - "#Create two subplots on 1 row and 2 columns with a figsize of (12, 8)\n", - "fig, ax = plt.subplots(___, ___, figsize=(___))\n", - "#Specify a horizontal barplot ('barh') as kind of plot (kind=)\n", - "ski_data.Region.value_counts().plot(kind=___, ax=ax[0])\n", - "#Give the plot a helpful title of 'Region'\n", - "ax[0].set_title(___)\n", - "#Label the xaxis 'Count'\n", - "ax[0].set_xlabel(___)\n", - "#Specify a horizontal barplot ('barh') as kind of plot (kind=)\n", - "ski_data.state.value_counts().plot(kind=___, ax=ax[1])\n", - "#Give the plot a helpful title of 'state'\n", - "ax[1].set_title(___)\n", - "#Label the xaxis 'Count'\n", - "ax[1].set_xlabel(___)\n", - "#Give the subplots a little \"breathing room\" with a wspace of 0.5\n", - "plt.subplots_adjust(wspace=___);\n", - "#You're encouraged to explore a few different figure sizes, orientations, and spacing here\n", - "# as the importance of easy-to-read and informative figures is frequently understated\n", - "# and you will find the ability to tweak figures invaluable later on" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "How's your geography? Looking at the distribution of States, you see New York accounting for the majority of resorts. Our target resort is in Montana, which comes in at 13th place. You should think carefully about how, or whether, you use this information. Does New York command a premium because of its proximity to population? Even if a resort's State were a useful predictor of ticket price, your main interest lies in Montana. Would you want a model that is skewed for accuracy by New York? Should you just filter for Montana and create a Montana-specific model? This would slash your available data volume. Your problem task includes the contextual insight that the data are for resorts all belonging to the same market share. This suggests one might expect prices to be similar amongst them. You can look into this. A boxplot grouped by State is an ideal way to quickly compare prices. Another side note worth bringing up here is that, in reality, the best approach here definitely would include consulting with the client or other domain expert. They might know of good reasons for treating states equivalently or differently. The data scientist is rarely the final arbiter of such a decision. But here, you'll see if we can find any supporting evidence for treating states the same or differently." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.3.5 Distribution Of Ticket Price By State" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Our primary focus is our Big Mountain resort, in Montana. Does the state give you any clues to help decide what your primary target response feature should be (weekend or weekday ticket prices)?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.3.5.1 Average weekend and weekday price by state" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 14#\n", - "# Calculate average weekday and weekend price by state and sort by the average of the two\n", - "# Hint: use the pattern dataframe.groupby()[].mean()\n", - "state_price_means = ski_data.___(___)[[___, ___]].mean()\n", - "state_price_means.head()" - ] - }, - { - "cell_type": "code", - "execution_count": 18, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqAAAAJcCAYAAAAmbBanAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeZgX1Zn+//eddgHD5kIcRWO7IKjQAWxQEA0uURONGtGvGhNFowSTicGMmfgzo2kdjSQxi2ImBEfpLAQTURBwY1RQA4g2NDYIbqOoSFQgioA0I/D8/qjT7Ye2F9ZPb/fruvrqqnNOnXqqmlx5PKeqjiICMzMzM7N8+UxjB2BmZmZmrYsTUDMzMzPLKyegZmZmZpZXTkDNzMzMLK+cgJqZmZlZXjkBNTMzM7O8cgJqZtbESHpY0sWb0W6xpJPyEVNDJI2SdN127rNE0p+3Z59m1jQ4ATWzZkfSdEnvS9q1sWPZVrUlWRHx5Yj4ww4853ZPXCNiWET85/bsc3uSNFDSTEkrJf1T0gxJfVPdEEl/34K+CiWFpJ12XMRmLZsTUDNrViQVAscCAZyxA/p3UrGFJBU0dgz1kdQBmAKMBPYAugA3AOsaMy6z1swJqJk1NxcBzwClwMUAknaV9IGkHlWNJHWWtFbS59L+6ZLmpXYzJRXltF0s6UeSKoA1knaSdI2k/5W0StJCSV/LaV8g6ZeSlkt6XdK/5o6ISeoo6S5J/5D0tqSbakvSJJ0KXAucJ2m1pOdT+XRJl+W0u1zSopxY+tTSV/cUy/n1Xa+kPwGfByanc/57LX0NkrRE0rXpGhdLujCnvlTS7yQ9JGkNcHwquymnzZnp/B+m+3jqltybHG0k/TVd+1xJX0j9/FDSfTXiHinpN7X0cShARIyLiA0RsTYipkZEhaTDgFFA/3Q/Pkh9nSapPMX/lqSSnP6eSr8/SMf0T8dcmv5O70t6VNIB9VyXWesWEf7xj3/802x+gFeB7wBHAh8De6fyu4Gbc9p9F3gkbfcB3gOOAgrIEtfFwK6pfjEwD9gfaJvKzgX2JfsP9fOANcA+qW4YsBDYD9gdeIxsRHanVD8R+D3wWeBzwLPAt+u4nhLgzzXKpgOX5cTxNtAXEHAIcEBO3Cel63sTOH0Lrvekeu7xIGA98CtgV+CL6fq7pfpSYCVwTLo/bVLZTam+X6r/UqrvAnTfynvzMXAOsDNwNfB62t4nxdQptd0pXfORtfTTAVgB/AH4MrB7jfohwN9ruQc9U/xFwLvAWamuMPfvncrOIvu3eViK5T+AmY39vxf/+Kep/ngE1MyaDUkDgQOAv0XEHOB/ga+n6r8AF+Q0/3oqA7gc+H1EzI5sBOwPZNOvR+e0vz0i3oqItQARcW9ELI2IjRHxV+AVssQK4P8Bt0XEkoh4HxiRE+PeZEnO8IhYExHvAb8Gzt/Ky74M+HlEPBeZVyPijZz6Y4FJwMURMWULrndzXBcR6yLiSeBBsuuu8kBEzEj3p7LGcd8C7o6I/0n1b0fEi1t5b+ZExPiI+JgsIW4DHB0R/yAbiTw3tTsVWJ7+XWwiIj4EBpIljXcCyyRNSvHUKiKmR8T8FH8FMI4sEa/Lt4FbImJRRKwHfgr08iioWe2cgJpZc3IxMDUilqf9v6QygCeAtpKOSv+n3wuYkOoOAP4tTUd/kKZZ9ycb4azyVu6JJF2UM4X9AdAD2CtV71ujfe72AWQjdP/IOfb3ZKN9W2N/skS7LsPIRtqm1YihoettyPsRsSZn/w3quV+bGfPW3Jvq80TERmBJThx/AL6Rtr8B/KmuTlJiOCQi9iP7W+4L1DZdD0D6dzRN0jJJK8nu8151tU/XdlvOdf2TbMS6Sz3HmLVaftjezJoFSW3JRuAKJL2TincFOkn6QkQ8L+lvZKOg7wJTImJVavcW2fT8zfWcInLOdQDZSNmJwKyI2CBpHllCAfAPsun3KvvnbL9FNtq4VxoJa0g0UP8WcHA99cOAH0n6dURclXNMfdfb0DkBdpf02Zwk9PPAgs3so66Yt/TeQM69lfQZsvu+NBVNBH6n7Nnf04FPPc9amzQaW0o2agm1X8tfgDuAL0dEZXq2dK962lfd87GbE4NZa+cRUDNrLs4CNgCHk41u9iJ73u5psheTIEsazgMu5JPpd8iSyWFpVEuSPpteMmlfx7k+S5ZkLAOQdAnZqFmVvwHfl9RFUifgR1UVaWp4KvBLSR0kfUbSwZLqmr59FyhMyVVt/hu4WtKRKfZDakzrriKbfj5OUtWjAA1d77vAQXWcL9cNknaRdCxZgnfvZhwDcBdwiaQT0/V3kdR9K+4NwJGSzlb2gtdwsgT2GYA09T+e7G/9bES8WVsHyl7Q+jdJ+6X9/cn+Q+WZ1ORdYD9Ju+Qc1h74Z0o++/HJox6Q/bvYyKb3cBTw/0k6Ip2jo6RzMbNaOQE1s+biYmBMRLwZEe9U/ZCNUl0oaaeImE32Ysq+wMNVB0ZEGdlzkXcA75O9LDKkrhNFxELgl8AssuSkJzAjp8mdZIlUBVAOPET20s6GVH8RsAvZi0rvkyVJ+9RxuqqkboWkubXEci9wM1mStYps1G+PGm0+IHvh58uS/nMzrvcW4D/SdPHVdcT1Tjp2KTAWGBYRL9bRtmbMzwKXkD3fuRJ4kmyKGrbs3gA8QPYfFe8D3wTOTs+DVvkD2d+nzul3svt2FDBb2Vv7z5CN5v5bqn8CeAF4R1LV4x3fAW6UtAq4nuw/Oqqu7yOyv8mMdA+PjogJwM+AeyR9mPr/cj0xmbVqiticmRgzM6uLpC8DoyKiRbxwImkQ2Zv5+zXUtrFJ+jzwIvAv6WUjM2sGPAJqZraFJLWV9BVl3wvtAvyET154sjxJjy38ALjHyadZ8+KXkMzMtpzIVtL5K7CW7BNF1zdqRK2MpM+SPR7xBtkzsGbWjHgK3szMzMzyylPwZmZmZpZXnoJvRvbaa68oLCxs7DDMzMzMGjRnzpzlEdG5tjonoM1IYWEhZWVljR2GmZmZWYMkvVFXnafgzczMzCyvnICamZmZWV45ATUzMzOzvGpyz4BK2gDMJ4vtdeCbEfFBWpnj6og4fSv73aLjJRUCUyKiRz1tVkdEu62JZ0vPZWZm1pp8/PHHLFmyhMrKysYOxRrQpk0b9ttvP3beeefNPqbJJaDA2ojoBSDpD8B3ydbcNTMzs1ZiyZIltG/fnsLCQiQ1djhWh4hgxYoVLFmyhAMPPHCzj2vqU/CzgC45++0kjZf0oqSxSv8iJZ0oqVzSfEl3S9o1lZ+a2v4dOLuqE0mfTe2eS8edWV8Qko6Q9KykeZIqJHWtUd9O0uOS5qYYzkzlhZIWSbpT0guSpkpqm+qOlPS8pFlkSbaZmZkllZWV7Lnnnk4+mzhJ7Lnnnls8Ut1kE1BJBcCJwKSc4t7AcOBw4CDgGEltgFLgvIjoSTaqe0UqvxP4KnAs8C85/fwYeCIi+gLHA79Iy7rVZRhwWxqZLQaW1KivBL4WEX1Sf7/UJ/+L6Qr8NiKOAD4ABqfyMcCVEdG/gfswVFKZpLJly5bV19TMzKxFcfLZPGzN36kpJqBtJc0DVgB7AP+TU/dsRCyJiI3APKAQ6Aa8HhEvpzZ/AI4DuqfyVyJbb/TPOf2cDFyTzjMdaAN8vp6YZgHXSvoRcEBErK1RL+CnkiqAx8hGbfdOda9HxLy0PQcolNQR6BQRT6byP9V14ogYHRHFEVHcuXOt33I1MzMza1aa7DOgKUmbQjY9fXuqW5fTbgNZ/PWl3XUtdC9gcES8tDkBRcRfJM0GTgMelXRZRDyR0+RCoDNwZER8LGkxWVJbW8xt0/nris3MzMxqKLzmwe3a3+IRp21WuwkTJnD22WezaNEiunfv/qn6QYMGceutt1JcXFxnHyUlJbRr146rr76a0tJSTj75ZPbdd18eeOABxowZw8SJEwG45ZZbuOuuu3j11VcBmDx5MnfeeSeTJk2qs+/alJaWUlZWxh133FFvu3bt2rF69eot6nt7aYojoABExErgSuBqSfW9VvUi2ajiIWn/m8CTqfxASQen8gtyjnkU+F7OM6S964tF0kHAaxFxO9kjAUU1mnQE3kvJ5/HAAQ1c2wfASkkDU9GF9bU3MzOzxjFu3DgGDhzIPffcs136Ky0tZenSpQAMGDCAWbNmVdfNmjWLDh068N577wEwc+ZMjjnmmO1y3qamySagABFRDjwPnF9Pm0rgEuBeSfOBjcCoVD4UeDC9hJS7HNR/AjsDFZIWpP36nAcsSFP23YE/1qgfCxRLKiNLJl/cjMu7BPhtegmp5pS+mZmZNbLVq1czY8YM7rrrruoEdO3atZx//vkUFRVx3nnnsXbtJ/8X3q7dJ19mHD9+PEOGDNmkv/Hjx1NWVsaFF15Ir169aNeuHR07dqwe8Xz77bcZPHgwM2fOBLIEdMCAASxbtozBgwfTt29f+vbty4wZMwBYs2YNl156KX379qV379488MADn7qGBx98kP79+7N8+XJef/11+vfvT9++fbnuuus2uc4TTzyRPn360LNnz+p+rrvuOm677bbqdj/+8Y+5/fbbP3WOrdHkpuBrflczIr6aszs9p/xfc7YfJ3tBqWZfj5AljDXL1wLfbiCOxUCPtH0LcEtdsUbEcqCul4l65LS/NWd7DvCFnHYl9cVjZmZm+TVx4kROPfVUDj30UPbYYw/mzp3L9OnT2W233aioqKCiooI+ffpsdn/nnHMOd9xxxyZT9gMGDGDmzJls2LCBrl27cvTRR/Poo49y+umnU1FRQd++fbn00ku56qqrGDhwIG+++SannHIKixYt4uabb+aEE07g7rvv5oMPPqBfv36cdNJJ1eebMGECv/rVr3jooYfYfffdufTSS7niiiu46KKL+O1vf1vdrk2bNkyYMIEOHTqwfPlyjj76aM444wy+9a1vcfbZZ/P973+fjRs3cs899/Dss89ul3vb5BJQMzMzs6Zg3LhxDB8+HIDzzz+fcePG8corr3DllVcCUFRURFFRzafytswxxxxTnYD279+ffv36ceONN1JeXk63bt1o06YNjz32GAsXLqw+5sMPP2TVqlVMnTqVSZMmceut2fhWZWUlb775JgDTpk2jrKyMqVOn0qFDBwBmzJjBfffdB8A3v/lNfvSjHwHZtzyvvfZannrqKT7zmc/w9ttv8+6771JYWMiee+5JeXk57777Lr1792bPPffcpuut4gTUzMzMrIYVK1bwxBNPsGDBAiSxYcMGJNG7d+86PzuUW76538UcMGAAI0eOZMOGDVx++eW0b9+eyspKpk+fXv3858aNG5k1axZt27bd5NiI4L777qNbt26blM+ePZuDDjqI1157jZdffnmTF6Rqi33s2LEsW7aMOXPmsPPOO1NYWFgd/2WXXUZpaSnvvPMOl1566WZd0+Zo0s+AmpmZmTWG8ePHc9FFF/HGG2+wePFi3nrrLQ488ED69OnD2LFjAViwYAEVFRXVx+y9994sWrSIjRs3MmHChFr7bd++PatWrareP/zww1m6dClPP/00vXtnTxP26tWLUaNGMWDAAABOPvnkTd5onzcv+7rjKaecwsiRI8m+Ngnl5eXVbQ444ADuv/9+LrroIl544QUgG22tepa16hoAVq5cyec+9zl23nlnpk2bxhtvfPLazNe+9jUeeeQRnnvuOU455ZQtvY118giomZmZNXmb+9mk7WXcuHFcc801m5QNHjyY8vJy1q5dS1FREb169aJfv37V9SNGjOD0009n//33p0ePHrV+4mjIkCEMGzaMtm3bVo9qHnXUUaxcubJ6LfX+/fszevTo6gT09ttv57vf/S5FRUWsX7+e4447jlGjRnHdddcxfPhwioqKiAgKCwuZMmVK9bm6devG2LFjOffcc5k8eTK33XYbX//617ntttsYPHhwdbsLL7yQr371qxQXF9OrV69NPje1yy67cPzxx9OpUycKCgq2z80FVJU1W9NXXFwcZWVljR2GmZnZDrdo0SIOO+ywxg6j1du4cSN9+vTh3nvvpWvXrnW2q+3vJWlORNT6gVRPwZuZmZnZpyxcuJBDDjmEE088sd7kc2t4Cn4rSVpd85NRO9zScijpmNdTmtk2KlnZ2BGYmW2Vww8/nNdee22H9O0RUDMzMzPLKyeg20iZX0haIGm+pPNS+X9JOiNtT5B0d9r+lqSb0vZESXMkvSBpaONdhZmZmVn+eAp+250N9CJb1Wgv4DlJTwFPAceSrR3fBdgntR8IVC0oe2lE/FNS23TcfRGxIrfzlJgOBSjo0JnCyjE7+nrMWoR8vzFrZmabzyOg224gMC4iNkTEu8CTQF/gaeBYSYcDC4F3Je1DtmTnzHTslZKeB54B9gc+9YRvRIyOiOKIKC7Yzc9/mpmZWfPnEdBtV+tyCBHxtqTdgVPJRkP3AP4fsDoiVkkaBJwE9I+IjyRNB9rkJ2QzM7NmZnu/hLuZLwhOmDCBs88+m0WLFm3yfcwqgwYN2mRt91pPVVJCu3btuPrqqyktLeXkk09m33335YEHHmDMmDFMnDgRgFtuuYW77rqLV199FYDJkydz5513MmnSpC26tNLSUsrKyjb5eP32UFhYSFlZGXvttdc29+UR0G33FHCepAJJnYHjgGdT3SxgeGrzNHB1+g3QEXg/JZ/dgaPzG7aZmZk1ZNy4cQwcOLB6BaFtVVpaytKlS4FsGc5Zs2ZV182aNYsOHTrw3nvvATBz5szq5ThbGo+AbrsJZNPqzwMB/HtEvJPqngZOjohXJb1BNgpalYA+AgyTVAG8RDYNX6+eXTpS5ufazMzM8mL16tXMmDGDadOmccYZZ1BSUsLatWu55JJLWLhwIYcddhhr166tbt+uXbvq1Y/Gjx/PlClTKC0tra4fP348ZWVlXHjhhdUrIXXs2JFXX32VQw45hLfffpvBgwczc+ZMzjrrLGbOnMlNN93EsmXLGDZsGG+++SYAv/nNbzjmmGNYs2YN3/ve95g/fz7r16+npKSEM888c5NrePDBB7npppuYPHkyc+fO5Sc/+Qnr1q3j4IMPZsyYMbRr147CwkIuvvhiJk+ezMcff8y9995L9+7dWbFiBRdccAHLli2jX79+bM/FizwCupWqvgEamR9GRI+I6BkRf81pc1dE7Ju2P46Iz0bE/Wl/XUR8OSKKIuLciBgUEdMb5WLMzMzsUyZOnMipp57KoYceyh577MHcuXP53e9+x2677UZFRQU//vGPmTNnzmb3d84551BcXMzYsWOZN28ebdu2ZcCAAcycOZOXXnqJrl27cvTRRzNz5kzWr19PRUUFffv25fvf/z5XXXUVzz33HPfddx+XXXYZADfffDMnnHACzz33HNOmTeOHP/wha9asqT7fhAkTGDFiBA899BAAN910E4899hhz586luLiYX/3qV9Vt99prL+bOncsVV1zBrbfeCsANN9zAwIEDKS8v54wzzqhOgLcHj4CamZmZ1WLcuHEMHz4cgPPPP59x48bxyiuvcOWVVwJQVFREUVHRNp3jmGOOYebMmWzYsIH+/fvTr18/brzxRsrLy+nWrRtt2rThscceY+HChdXHfPjhh6xatYqpU6cyadKk6oSxsrKyOkmcNm0aZWVlTJ06lQ4dOjBlyhQWLlxYPaX/f//3f/Tv37+6z7PPPhuAI488kvvvvx+Ap556qnr7tNNOY/fdd9+ma83lBNTMzMyshhUrVvDEE0+wYMECJLFhwwYk0bt3b6Ra3z/epLyysnKzzjNgwABGjhzJhg0buPzyy2nfvj2VlZVMnz69OlncuHEjs2bNom3btpscGxHcd999dOvWbZPy2bNnc9BBB/Haa6/x8ssvU1xcTETwpS99iXHjxtUax6677gpAQUEB69evr/WatidPwZuZmZnVMH78eC666CLeeOMNFi9ezFtvvcWBBx5Inz59GDt2LAALFiygoqKi+pi9996bRYsWsXHjRiZMmFBrv+3bt2fVqlXV+4cffjhLly7l6aefpnfv3gD06tWLUaNGMWDAAABOPvnkTd5onzdvHgCnnHIKI0eOrH42s7y8vLrNAQccwP33389FF13ECy+8wNFHH82MGTOq37D/6KOPePnll+u9B8cdd1z1tT788MO8//77m3HnNo9HQM3MzKzp28zPJm0v48aN45prrtmkbPDgwZSXl7N27VqKioro1asX/fr1q64fMWIEp59+Ovvvvz89evSofiEp15AhQxg2bFj1S0ht27blqKOOYuXKley8884A9O/fn9GjR1cnoLfffjvf/e53KSoqYv369Rx33HGMGjWK6667juHDh1NUVEREUFhYyJQpU6rP1a1bN8aOHcu5557L5MmTKS0t5YILLmDdunVA9kzooYceWuc9+MlPfsIFF1xAnz59+OIXv8jnP//5rb+hNWh7vtFkO1ZxcXGUlZU1dhhmZmY73KJFizjssMMaOwzbTLX9vSTNiYhaP5DqKXgzMzMzyytPwTcnS8u3/0oQZtY05Hl60cysMXkEtBaSVqffhZIWpO1Bkqak7TMkXVNfH/X0va+k8dsvWjMzs5bJjwk2D1vzd3ICuhUiYlJEjNjKY5dGxDnbOyYzM7OWpE2bNqxYscJJaBMXEaxYsYI2bdps0XGegt8KkoYAxRHxr5JKgQ+BYuBfyJbiHK/sw1k/B75MtkTnTRHxV0mFwJSI6CHpCGAMsAvZfwwMjohX8n09ZmZmTc1+++3HkiVLWLZsWWOHYg1o06YN++233xYd4wR0+9gHGAh0ByYB44GzgV7AF4C9gOckPVXjuGHAbRExVtIuQEHNjiUNBYYCFHToTGHlmB12EWbN2eIRpzV2CGa2He28884ceOCBjR2G7SCegt8+JkbExohYCOydygYC4yJiQ0S8CzwJ9K1x3CzgWkk/Ag6IiLU1O46I0RFRHBHFBbv5BSQzMzNr/pyAbh/rcrZV43edIuIvwBnAWuBRSSfsgNjMzMzMmhQnoDvOU8B5kgokdQaOA57NbSDpIOC1iLidbOq+KP9hmpmZmeWXnwHdcSYA/YHnyV5C+veIeCe9hFTlPOAbkj4G3gFurK/Dnl06Uubn3MzMzKyZ81KczYiX4jQzM7PmwktxmpmZmVmT4QTUzMzMzPLKCaiZmZmZ5ZUTUDMzMzPLKyegZmZmZpZXTkDNzMzMLK/8HdDmZGk5lHg5TjMDSlY2dgRmZlvNI6DbQNJ+kh6Q9Iqk/5V0m6RdJA2RdEcdxzwkqVO+YzUzMzNrKpyAbiVJAu4HJkZEV+BQoB1wc33HRcRXIuKDPIRoZmZm1iQ5Ad16JwCVETEGICI2AFcBlwK7AftKeiSNjv686iBJiyXtlbZ/IGlB+hneCNdgZmZmlnd+BnTrHQHMyS2IiA8lvUl2X3sBvYF1wEuSRkbEW1VtJR0JXAIcBQiYLenJiCjP7VPSUGAoQEGHzhRWjtmBl2RmO9LiEac1dghmZk2CR0C3noCop/zxiFgZEZXAQuCAGu0GAhMiYk1ErCabzj+2ZmcRMToiiiOiuGA3v4BkZmZmzZ8T0K33AlCcWyCpA7A/sIFs5LPKBj492qwdGp2ZmZlZE+Up+K33ODBC0kUR8UdJBcAvgVLgo804/imgVNIIsmT0a8A36zugZ5eOlHkKz8zMzJo5j4BupYgIsqTxXEmvAC8DlcC1m3n8XLJk9VlgNvDfNZ//NDMzM2uJlOVR1hwUFxdHWVlZY4dhZmZm1iBJcyKiuLY6j4CamZmZWV45ATUzMzOzvHICamZmZmZ55QTUzMzMzPLKCaiZmZmZ5ZW/A9qcLC2HEq+GZGbboGRlY0dgZtb6RkAlhaQ/5ezvJGmZpClp/wxJ1zRCXMMkXZTv85qZmZnlW2scAV0D9JDUNiLWAl8C3q6qjIhJwKR8BiRpp4gYlc9zmpmZmTWWVjcCmjwMVK1peQEwrqpC0hBJd6TtcyUtkPS8pKdS2RGSnpU0T1KFpK6p/Aep7QJJw1NZoaQFOX1fLakkbU+X9FNJTwLfl1Qi6eo8XLuZmZlZo2qNI6AA9wDXp2n3IuBu4Nha2l0PnBIRb0vqlMqGAbdFxFhJuwAFko4ELgGOIlvXfXZKLN9vII5OEfFFgKrEtCZJQ4GhAAUdOlNYOWYLLtPMmrrFI05ruJGZWQvTKkdAI6ICKCQb/XyonqYzgFJJlwMFqWwWcK2kHwEHpGn8gcCEiFgTEauB+6k9oa3pr5sR6+iIKI6I4oLd/AKSmZmZNX+tMgFNJgG3kjP9XlNEDAP+A9gfmCdpz4j4C3AGsBZ4VNIJZKOetVnPpve4TY36NVsZu5mZmVmz1ZoT0LuBGyNifl0NJB0cEbMj4npgObC/pIOA1yLidrIktgh4CjhL0m6SPgt8DXgaeBf4nKQ9Je0KnL6Dr8nMzMysyWutz4ASEUuA2xpo9ov0kpGAx4HngWuAb0j6GHiHLIn9p6RS4Nl03H9HRDmApBuB2cDrwIvbEnPPLh0p8/NiZmZm1swpIho7BttMxcXFUVZW1thhmJmZmTVI0pyIKK6trjVPwZuZmZlZI3ACamZmZmZ55QTUzMzMzPLKCaiZmZmZ5ZUTUDMzMzPLKyegZmZmZpZXrfY7oM3S0nIo8XKcZradlKxs7AjMrJVq9SOgklZvYftBkqak7TMkXbNjIjMzMzNrmTwCug0iYhLZcpxmZmZmtpla/QholTSyOV3SeEkvShorSanu1FT2d+DsnGOGSLojbX9V0mxJ5ZIek7R3Ki+RdHfq+zVJV+YcP1HSHEkvSBqa50s2MzMzaxQeAd1Ub+AIYCkwAzhGUhlwJ3AC8Crw1zqO/TtwdESEpMuAfwf+LdV1B44H2gMvSfpdRHwMXJrWkW8LPCfpvohYkdtpSkyHAhR06Exh5ZjteLlm1pwsHnFaY4dgZrZdOAHd1LMRsQRA0jygEFgNvB4Rr6TyP5MSwhr2A/4qaR9gF+D1nLoHI2IdsE7Se8DewBLgSklfS232B7oCmySgETEaGA2w6z5dY3tcpJmZmVlj8hT8ptblbG/gkwR9cxK/kcAdEdET+DbQpr5+JQ0CTgL6R8QXgPIax5iZmZm1SE5AG/YicKCkg9P+BXW06wi8nbYv3kjRXjoAACAASURBVIx+OwLvR8RHkroDR29bmGZmZmbNg6fgGxARlek5zAclLSd71rNHLU1LgHslvQ08AxzYQNePAMMkVQAvpWPq1bNLR8r8DJiZmZk1c4rwY4XNRXFxcZSVlTV2GGZmZmYNkjQnIoprq/MUvJmZmZnllRNQMzMzM8srJ6BmZmZmlldOQM3MzMwsr5yAmpmZmVle+TNMzcnScijp2NhRmFlrULKysSMwsxasRYyASgpJf8rZ30nSMklT0v4Zkq5pvAhB0iBJAxozBjMzM7OmoKWMgK4BekhqGxFrgS/xyapERMQkYFJjBZcMIltXfmYjx2FmZmbWqFrECGjyMFC1TNAFwLiqCklDJN2Rtksl3S5ppqTXJJ2TygdJmi5pvKQXJY2VpFR3pKQnJc2R9KikfVL5lZIWSqqQdE8q20PSxFT2jKQiSYXAMOAqSfMkHSvpq5JmSyqX9JikvfN0n8zMzMwaVUsZAQW4B7g+TbsXAXcDx9bRdh9gINCdbGR0fCrvDRwBLAVmAMdImg2MBM6MiGWSzgNuBi4FrgEOjIh1kjqlPm4AyiPiLEknAH+MiF6SRgGrI+JWAEm7A0dHREi6DPh34N9qBpqWAR0KUNChM4WVY7b2/phZM7XYS/CaWQvTYhLQiKhII40XAA810HxiRGwEFtYYeXw2IpYASJoHFAIfkK39/j9pQLQA+EdqXwGMlTQRmJjKBgKDU0xPSNpTUm1vDu0H/DWNpu4CvF7HdY0GRgPsuk9Xr5tqZmZmzV5LmoKHbDTzVnKm3+uwLmdbdZRvIEvQBbwQEb3ST8+IODm1OQ34LXAkMEfSTjX6q1Jb4jgSuCMiegLfBto0ELOZmZlZi9DSEtC7gRsjYv527PMloLOk/gCSdpZ0hKTPAPtHxDSy6fNOQDvgKeDC1HYQsDwiPgRWAe1z+u3IJy9KXbwd4zUzMzNr0lrMFDxAmj6/bTv3+X/pRaXb01T6TsBvgJeBP6cyAb+OiA8klQBjJFUAH/FJcjkZGC/pTOB7QAlwr6S3gWeAAxuKpWeXjpT5WTAzMzNr5hThxwqbi+Li4igrK2vsMMzMzMwaJGlORBTXVtfSpuDNzMzMrIlzAmpmZmZmeeUE1MzMzMzyygmomZmZmeWVE1AzMzMzyysnoGZmZmaWVy3qO6At3tJyKKltVU8zsyaoZGVjR2BmTVSLHwGVFJL+lLO/k6Rlkqak/TMkXVPP8YWSFtRRd6Okk7Z/1GZmZmYtV2sYAV0D9JDUNiLWAl/ikyUwiYhJZGvIb7GIuH77hGhmZmbWerT4EdDkYaBqDcsLgHFVFZKGSLojbe8taYKk59PPgNSsQNKdkl6QNFVS29S+NC3TiaSvSHpR0t8l3Z4zwtpP0kxJ5el3t5zz3i/pEUmvSPp5fm6FmZmZWeNqDSOgAPcA16eksAi4Gzi2lna3A09GxNckFQDtgN2BrsAFEXG5pL8Bg4E/Vx0kqQ3we+C4iHhd0ricPl9M5evTdP1P0/EAvYDewDrgJUkjI+Kt3IAkDQWGAhR06Exh5ZhtuhFm1jIsHnFaw43MzJqoVjECGhEVQCHZ6OdD9TQ9AfhdOmZDRFQ9Qf96RMxL23NSX7m6A69FxOtpPzcB7Qjcm54j/TVwRE7d4xGxMiIqgYXAAbXEPjoiiiOiuGA3v4BkZmZmzV+rSECTScCtbJocbq51Odsb+PTIseo59j+BaRHRA/gq0GYL+jUzMzNrcVpTAno3cGNEzK+nzePAFQCSCiR12My+XwQOklSY9s/LqevIJy89DdncYM3MzMxaqlYz4hYRS4DbGmj2fWC0pG+RjUheAfxjM/peK+k7wCOSlgPP5lT/HPiDpB8AT2xV8EnPLh0p83NfZmZm1swpIho7hhZBUruIWC1JwG+BVyLi19vzHMXFxVFWVrY9uzQzMzPbISTNiYji2upa0xT8jna5pHnAC2TT7r9v5HjMzMzMmqRWMwW/o6XRzu064mlmZmbWEnkE1MzMzMzyygmomZmZmeWVE1AzMzMzyys/A9qcLC2HEq+GZGa2iZKVDbcxsyYlryOgkkLSn3L2d5K0LK3RnneSBu3Ic0vqlL4PWrVfKOnrO+p8ZmZmZs1Bvqfg1wA9JLVN+1/ik1WCWqJOwHdy9gsBJ6BmZmbWqjXGM6APA1XL+VxAztrskvpJmimpPP3ulsqPkPSspHmSKiR1lfRZSQ9Kel7SAknnpbbXS3oulY1OH4ZH0iGSHkvt50o6OJ22naTxkl6UNDan/WJJe6XtYknT0/YXUxzzUpztU/kP03krJN2Q+h4BHJza/iLtH5v2r6rtunbYXTczMzNrIhrjGdB7gOvT1HcR2Rrtx6a6F4HjImK9pJOAnwKDgWHAbRExVtIuQAHwFWBpRJwGIKnq4cg7IuLGVPYn4HRgMjAWGBEREyS1IUu+9wd6A0cAS4EZwDHA3+uJ/2rguxExQ1I7oFLSyUBXoB8gYJKk44BrgB4R0SvFMwi4OiJOT/sja7muTUgaCgwFKOjQmcLKMQ3fYTNrNIu9XK6ZWYPyPgIaERVkU9EXAA/VqO4I3CtpAdlH3Y9I5bOAayX9CDggItYC84GTJP1M0rERUfUU+vGSZkuaD5wAHJFGKbtExIQUQ2VEfJTaPxsRSyJiIzAvxVafGcCvJF0JdIqI9cDJ6accmAt0J0tIG1LbdW0iIkZHRHFEFBfs5heQzMzMrPlrrM8wTQJuJWf6PflPYFpE9AC+CrQBiIi/AGcAa4FHJZ0QES8DR5Ilorekqfc2wH8B50RET+DO1IfqiWVdzvYGPhkVXs8n96dNVYOIGAFcBrQFnpHUPfV/S0T0Sj+HRMRdDd2E2q6roWPMzMzMmrvGSkDvBm6MiPk1yjvyyUtJQ6oKJR0EvBYRt5Mlr0WS9gU+iog/kyWzffgkUVyepsfPAYiID4Elks5K/e0qabcGYlxMluBC9hhAVSwHR8T8iPgZUEY22vkocGk6J5K6SPocsApon9PnJvu1XVcDMZmZmZk1e43yHdCIWALcVkvVz4E/SPoB8ERO+XnANyR9DLwD3Aj0BX4haSPwMXBFRHwg6U6yUdHFwHM5fXwT+L2kG1P7cxsI8wbgLknXArNzyodLOp5stHQh8HBErJN0GDArvcO0GvhGRPyvpBnpkYKHgWuB9ZKeB0rJEuaa11Wnnl06Uubny8zMzKyZU0Q0dgy2mYqLi6OsrKyxwzAzMzNrkKQ5EVFcW52X4jQzMzOzvHICamZmZmZ55QTUzMzMzPLKCaiZmZmZ5ZUTUDMzMzPLKyegZmZmZpZXjfIdUNtKS8uhxMtxmlkLUrKy4TZm1uK0mBFQSb+WNDxn/1FJ/52z/8u0XOc1jRTfvpLGN8a5zczMzJqSFpOAAjOBAQCSPgPsBRyRUz8AeDSt5Z53EbE0Is5pjHObmZmZNSUtKQGdQUpAyRLPBcAqSbtL2hU4DPiCpDsAJJ0raYGk5yU9lcoKJN0qab6kCknfS+UnSipP5Xen/pC0WNINkuamuu6p/IuS5qWfckntJRWmJTmRNETS/ZIekfSKpJ/n80aZmZmZNaYW8wxoRCyVtF7S58kS0VlAF6A/sBKoAP4v55DrgVMi4m1JnVLZUOBAoHdErJe0h6Q2ZOu2nxgRL0v6I3AF8Jt0zPKI6CPpO8DVwGXp93cjYoakdkBlLSH3AnoD64CXJI2MiLdqNpI0NMVFQYfOFFaO2bobZGYt3uIRpzV2CGZmm6UljYDCJ6OgVQnorJz9mbW0LZV0OVCQyk4CRkXEeoCI+CfQDXg9Il5Obf4AHJfTz/3p9xygMKfvX0m6EuhU1V8Nj0fEyoioBBYCB9R2QRExOiKKI6K4YDe/gGRmZmbNX0tLQKueA+1JNgX/DNkI6ACypLBaRAwD/gPYH5gnaU9AQNToUw2cc136vYE0opyeM70MaAs8UzU1X8dxmxxrZmZm1tK1tAR0BnA68M+I2JBGMDuRJaGzchtKOjgiZkfE9cByskR0KjBM0k6pzR7Ai0ChpEPSod8EnqwviNT3/Ij4GVAG1JaAmpmZmbVKLW3UbT7Z2+9/qVHWLiKWS5sMZv5CUleyEc7HgefJRk0PBSokfQzcGRF3SLoEuDclps8BoxqIY7ik48lGNhcCDwP7bOvF9ezSkTI/42VmZmbNnCJqzjhbU1VcXBxlZWWNHYaZmZlZgyTNiYji2upa2hS8mZmZmTVxTkDNzMzMLK+cgJqZmZlZXjkBNTMzM7O8cgJqZmZmZnnV0j7D1LItLYcSr4ZkZtbslaxs7AjMGlWzT0DTCkaPp91/Ifv25rK03y8i/q/WA83MzMysUTT7BDQiVgC9ACSVAKsj4tZGDcrMzMzM6tQinwGVdKSkJyXNkfSopH1S+XRJP5P0rKSXJR2byodIul/SI5JekfTznL5OljRL0lxJ90pql8pHSFooqULSransXEkLJD0v6alUViDpF5KeS22/ndP3D3PKb8jnPTIzMzNrLM1+BLQWAkYCZ0bEMknnATcDl6b6nSKin6SvAD8BTkrlvYDewDrgJUkjgbXAfwAnRcQaST8CfiDpDuBrQPeICEmdUh/XA6dExNs5Zd8CVkZEX0m7AjMkTQW6pp9+KeZJko6LiKc2uRhpKDAUoKBDZworx2y/O2XWCiz28rVmZk1OS0xAdwV6AP+T1n4vAP6RU39/+j0HKMwpfzwiVgJIWggcAHQCDidLGgF2AWYBHwKVwH9LehCYkvqYAZRK+lvOeU4GiiSdk/Y7kiWeJ6ef8lTeLpVvkoBGxGhgNMCu+3T1uqlmZmbW7LXEBFTACxHRv476den3Bja9/nU521V1Av4nIi741EmkfsCJwPnAvwInRMQwSUcBpwHzJPVKfXwvIh6tcfwpwC0R8fstvUAzMzOz5qwlPgO6DugsqT+ApJ0lHbGVfT0DHCPpkNTXbpIOTc+BdoyIh4DhfPIS1MERMTsirgeWA/sDjwJXSNo5tTlU0mdT+aU5z5R2kfS5rb1oMzMzs+aiJY6AbgTOAW6X1JHsGn8DvLClHaVnSIcA49Lzm5A9E7oKeEBSG7IRzqtS3S8kdU1ljwPPAxVkU/1zlc3jLwPOioipkg4DZqXp/dXAN4D36oqnZ5eOlPl5NjMzM2vmFOHHCpuL4uLiKCsra+wwzMzMzBokaU5EFNdW1xKn4M3MzMysCXMCamZmZmZ55QTUzMzMzPLKCaiZmZmZ5ZUTUDMzMzPLKyegZmZmZpZXLfE7oC3X0nIo6djYUZiZfaJkZWNHYGbNUJMZAZX0Y0kvSKqQNC8tabk1/QySNCBnvzRnHfb6jtuQzvuCpOcl/UBSvfcnnWtKfW1qtB8uabfNbW9mZmbWEjWJEdC0bObpQJ+IWCdpL2CXrexuENmqQjO38Li1EVG1pObngL8AHYGfbGUctRkO/Bn4aDv2aWZmZtasNJUR0H2A5RGxDiAilkfEUgBJJ0oqlzRf0t1VS2JKWpwSVSQVS5ouqRAYBlyVRjOPTf0fJ2mmpNc2ZzQ0It4DhgL/qkyhpKclzU0/A2oeI6lvivOg2mKWdCWwLzBN0rR0zO8klaVR1xu27RaamZmZNQ9NYgQUmApcL+ll4DHgrxHxZFprvRQ4MSJelvRH4Aqytd0/JSIWSxoFrI6IWwEkfYsswR0IdAcmAeMbCigiXktT8J8jW5/9SxFRmdZ6HwdULy2VEtKRwJmp7ZM1Y46I30j6AXB8RCxPh/44Iv4pqQB4XFJRRFTkxiFpKFkyTEGHzhRWjmkodDOzvFnc2AGYWbPUJEZAI2I1cCRZorUM+KukIUA34PWIeDk1/QNw3FacYmJEbIyIhcDeW3Cc0u+dgTslzQfuBQ7PaXMYMBr4akS8uYUx/z9Jc4Fy4Iga/QIQEaMjojgiigt28wtIZmZm1vw1lRFQImIDMB2YnhK9i4F59Ryynk8S6DYNdL8uZ1t1tsptJB0EbCAb0fwJ8C7whXTOypym/0jn7w0s3YL+DwSuBvpGxPuSSmn4OszMzMyavSYxAiqpW5rartILeAN4ESiUdEgq/ybZ9DZkMz9Hpu3BOceuAtpvYzydgVHAHRERZC8j/SMiNqYYCnKafwCcBvxU0qAGYs6NrQOwBlgpaW/gy9sSs5mZmVlz0VRGQNsBIyV1IhvZfBUYmp65vAS4V9JOwHNkiSHADcBdkq4FZuf0NRkYL+lM4HtbEENbSfPIptvXA38CfpXq/gu4T9K5wDSyxLFaRLwr6avAw8ClQF0xjwYelvSPiDheUjnwAvAaMKOhAHt26UjZiNO24JLMzMzMmh5lA3zWHBQXF0dZWVljh2FmZmbWIElzIqK4tromMQVvZmZmZq2HE1AzMzMzyysnoGZmZmaWV05AzczMzCyvnICamZmZWV41lc8w2eZYWg4lXg3JzJqokpWNHYGZNROtZgRUUkj6U87+TpKWSZqylf11kvSd7RehmZmZWevQahJQso/H95DUNu1/CXh7G/rrBDgBNTMzM9tCrSkBhWyloqqlhC4AxlVVSNpD0kRJFZKekVSUyksk3S1puqTXJF2ZDhkBHCxpnqRfSGon6XFJcyXNTysxIalQ0iJJd0p6QdLUqiRY0uWSnpP0vKT7JO2WtzthZmZm1kha2zOg9wDXp2n3IuBu4NhUdwNQHhFnSToB+CPZmvQA3YHjydZxf0nS74BrgB4R0QuyKX3gaxHxoaS9gGckTUrHdwUuiIjLJf2NbO36PwP3R8Sd6fibgG8BI3MDljQUGApQ0KEzhZVjtu8dMTPbThY3dgBm1my0qgQ0IiokFZKNfj5Uo3ogWWJIRDwhaU9JVW/8PBgR64B1kt4D9q6lewE/lXQcsBHoktPu9YiYl7bnAIVpu0dKPDsB7YBHa4l5NNka8uy6T1evm2pmZmbNXqtKQJNJwK3AIGDPnHLV0rYq4VuXU7aB2u/bhUBn4MiI+FjSYqBNHcdXPYdaCpwVEc9LGpJiMjMzM2vRWtszoJBNu98YEfNrlD9FlkQiaRCwPCI+rKefVWRT8lU6Au+l5PN44IDNiKU98A9JO1ed28zMzKyla3UjoBGxBLitlqoSYIykCuAj4OIG+lkhaYakBWQvN/0MmCypDJgHvLgZ4VwHzAbeAOazaUL7KT27dKRsxGn1NTEzMzNr8hThxwqbi+Li4igrK2vsMMzMzMwaJGlORBTXVtcap+DNzMzMrBE5ATUzMzOzvHICamZmZmZ55QTUzMzMzPLKCaiZmZmZ5ZUTUDMzMzPLq1b3HdBmbWk5lHRsuJ2ZWWtQsrKxIzCzrdRsRkAl/YukeyT9r6SFkh6SdGgjxjNc0m45+w9J6rQV/RRK+vr2jc7MzMys6WoWCagkAROA6RFxcEQcDlwL7N2IYQ0HqhPQiPhKRHywFf0UAk5AzczMrNVoFgkocDzwcUSMqiqIiHnA3yX9QtICSfMlnQfZWu6SpksaL+lFSWNTEoukxZJukDQ3HdM9lX9W0t2SnpNULunMVF4g6dbUtkLS9yRdCewLTJM0LaffvdL2Rant85L+lMpKJZ1TFb+k1WlzBHCspHmSrtqxt9HMzMys8TWXZ0B7AHNqKT8b6AV8AdgLeE7SU6muN3AEsBSYARwD/D3VLY+IPpK+A1wNXAb8GHgiIi5NU+nPSnoMuAg4EOgdEesl7RER/5T0A+D4iFieG5CkI1Jfx0TEckl7NHBt1wBXR8TptVVKGgoMBSjo0JnCyjENdGdmVrvFI05r7BDMzIDmMwJal4HAuIjYEBHvAk8CfVPdsxGxJCI2AvPIprqr3J9+z8kpPxm4RtI8YDrQBvg8cBIwKiLWA0TEPxuI6QRgfFViuhnt6xURoyOiOCKKC3bzC0hmZmbW/DWXEdAXgHNqKVc9x6zL2d7Apte6rpZyAYMj4qVNTpBN3ccWxFpX+/WkhD/1ucsW9GlmZmbWYjSXEdAngF0lXV5VIKkv8D5wXnpOszNwHPDsVp7jUeB7Oc+K9k7lU4FhknZK5f8/e/ceZ2dV33v882Wg4ZrBSkohBcfSIALRKFuO4Waw2J4KlquCh1YBa8RWKSj10GPFOXhDFLmDDZQEUAGRi1wsoNzvZEJCAgh6CrFIEIRoMEgiCd/zx7NGtsOeWzKzZ/bM9/16zWs/z7rvPf/8Xmut51ndS+q/ATZp0M5NwAckvb5H+cXATuV6X2C9ftqJiIiIGJNaYgbUtiXtD5wq6ThgBVVAdzSwMfAg1azjZ2z/ovvBokH6AnAqsLAEoYuBfYDzgG1L+svAucCZwCzgPyU9bXvPurE+LOlLwG2SVgPzgcNKve9Lup8qSH2xVFkIrJL0IDDH9im9DXDq5Ha6socrIiIiWpzswawux0iq1Wru6uoa6WFERERE9EvSPNu1RnmtsgQfEREREWNEAtCIiIiIaKoEoBERERHRVAlAIyIiIqKpEoBGRERERFMlAI2IiIiIpmqJ94A2W3nn6BXAm20/KqkDuNb2jmvQ1nLbGw/JwJbMh84cxxkR8Rqdy0Z6BBExCJkBbeyDwJ3AISM9kIiIiIixJgFoD5I2BnYFPkKDAFRSh6Q7JD1Q/nYp6VtIul3SAkkPSdq9R73NJN0jae/e2oiIiIgYD7IE/1r7Adfb/omkpZLeDiyty38WeI/tFZKmABcDNeB/ATfY/pKkNmDD7gqSNgeuBv7N9g8lbdhLG68haSYwE6Bt4iQ6Vswe8i8cEWPb4hzhGxGjTALQ1/og1ZnwAJeU+7Pq8tcDzpQ0DVhNdU48wFzgfEnrAVfZXlBX/ibgn2zf1k8br2F7FtW580zYYkrOTY2IiIiWlwC0jqTXA+8GdpRkoA0wcHZdsWOAZ4C3Um1hWAFg+3ZJewB7AxdJ+prtC4FVwDzgr4Hb+mojIiIiYjzIHtA/dBBwoe032O6wvRXwBPBndWXagadtvwL8PVWQiqQ3AM/aPhf4D+DtpbyBI4DtJB3XVxsRERER40FmQP/QB4ETe6RdDvyfuvuzgcslvR+4BXixpM8A/kXSy8By4EPdFWyvlnQIcI2kF/poo09TJ7fTlb1cERER0eJkZ1thq6jVau7q6hrpYURERET0S9I82w0fss4SfEREREQ0VQLQiIiIiGiqBKARERER0VQJQCMiIiKiqRKARkRERERTJQCNiIiIiKbKe0BbyZL50Nk+0qOIiIjB6Fw20iOIGHVGZAZUkiWdXHd/rKTOQbYxQ9IudfdzJB00gHp/KukSSf8l6RFJP5DU61nsgxhPh6SHynVN0ulr22ZERETEWDRSS/ArgQMkbbYmlSWtS3Xy0C79FO1ZT8CVwK22t7G9PdUpR5sPtL6kfn8z2122jxrM2CIiIiLGi5EKQFcBs4BjemZIeoOkmyQtLJ9bl/Q5kr4h6RbgUuBI4BhJCyTtXqrvIeluSY/3Mhu6J/Cy7W92J9heYPsOSRuX/h6QtEjSvqXfDkk/lnQ28ACwlaSvSXqolDu4wXeYIenact0p6XxJt5ZxHVVX7ipJ8yQ9LGnmmv2UEREREa1lJPeAngUslHRSj/QzgQttXyDpCOB0YL+Sty2wVzlbvRNYbvvrAJI+AmwB7AZsB1wNfK9H2zsC83oZzwpgf9svlJnZeyVdXfLeBBxu+x8lHQhMA94KbAbMlXR7P991O6rgdxPgMUnn2H4ZOML2UkkblHYut/18fcUSmM4EaJs4iY4Vs/vpKiJGm8Un7j3SQ4iIGFVG7Cl42y8AFwI9l6qnA98p1xdRBZTdLrO9uo9mr7L9iu1HGOCyeh0BX5a0EPgRMLmujZ/Zvrdc7wZcbHu17WeA24B39NP2dbZX2n4OeLau3aMkPQjcC2wFTOlZ0fYs2zXbtbYN8wBSREREtL6Rfg3TqcBHgI36KOO66xf7aW9l3bUa5D8M7NRL3UOBScBOtqcBzwDrN+i3Ubv9qR/XamBdSTOAvYDptt8KzK/rLyIiImLMGtEA1PZS4LtUQWi3u4FDyvWhwJ29VP8N1ZL2YNwMTJD00e4ESe+Q9C6gHXjW9suS9gTe0EsbtwMHS2qTNAnYA7h/kOOg9Pcr27+VtB3wzjVoIyIiIqLljIb3gJ4MfKLu/ijgfEn/AvwSOLyXetcA3ysPC31yIB3ZtqT9gVMlHUe173MxcDTV7Og1krqABcCjvTRzJdU2gQepZmc/Y/sXkjoGMoY61wNHliX/x6iW4fs0dXI7XdlLFhERES1OtvsvFaNCrVZzV1fXSA8jIiIiol+S5tmuNcob6T2gERERETHOJACNiIiIiKZKABoRERERTZUANCIiIiKaKgFoRERERDRVAtCIiIiIaKrR8B7QYSXJwDdsf7rcHwtsbLtzCNpen+qdoe+3vaikfQb4c9tHDqB+J3Xn2fdryXzozHGcERFrpHPZSI8gIorxMAO6EjhA0mZD3bDtFVQvsT9blcnAx4B/7a+upDEf/EdEREQ0Mh4C0FXALOCYnhmSJkm6XNLc8rdrSV8kadMSVD4v6UMl/SJJe9W3Yft64GngQ8ApQCcwUdJNkhaWz61L/TmSviHpFuCrPcbyUUn/KWmDIf8FIiIiIkaR8TILdxawUNJJPdJPA06xfWcJEm8A3gzcBewK/Ax4HNgduJDqvPaPN2j/aKrz4H9q+yJJ1wAX2r5A0hHA6cB+pey2wF62V5cleCR9AvgrYD/bK+sbljQTmAnQNnESHStmr8XPEBExOItz/G9EDINxEYDafkHShVTnzL9Ul7UXsL2k7vuJkjYB7gD2oApAzwFmluX1pbaXN2h/iaSbgWtL0nTggHJ9EVAf+F5me3Xd/d8DP6cKPl9u0PYsqhlcJmwxJeemRkRERMsbD0vw3U4FPgJsVJe2DjDd9rTyN9n2b4DbqWY9dwduBX4JHEQVmPbmlfLXSH3g+GKPvIeADuDPBvY1IiIiIlrbuAlAbS8FvksVhHa7EfhE942kaaXsk8BmwBTbjwN3AsfSdwBa727gkHJ9K5B2BAAAIABJREFUaKnfm/lUDy5dLWnLAbYfERER0bLGxRJ8nZOpCzipluTPkrSQ6re4Heh+fdJ9QFu5vgP4Cn0HkvWOAs6X9C9Us6eH91W47EE9FrhO0ntsP9eo3NTJ7XRlP1ZERES0ONnZVtgqarWau7q6RnoYEREREf2SNM92rVHeuFmCj4iIiIjRYUABqKQNJX1O0rnlfoqkfYZ3aBERERExFg10BnQ21YlC08v9z4EvDsuIIiIiImJMG2gAuo3tk4CXAWy/BKjvKhERERERrzXQAPR35YhIA0jahmpGNCIiIiJiUAb6GqZO4HpgK0nfpjqmss9XC0VERERENDLg1zBJej3VWegC7u3tXZUxfGpbtrlr5sYjPYyIiObrXDbSI4iIQVrr1zBJusn287avs32t7eck3TS0wxy9JL3m/PeSPkfSQf3UvVVSwx8/IiIiYjzqcwle0vrAhsBmkl7Hqw8eTQRybGREREREDFp/M6AfA+YB25XP7r/vA2cN79BGH1XOlPSIpOuAP6nLO17SXEkPSZolqf4tAe+XdL+kn0javZRfX9JsSYskzZe0Z7O/T0RERMRI6HMG1PZpwGmSPmn7jCaNaTTbH3gTMBXYHHgEOL/knWn7BABJFwH7ANeUvHVt7yzpvcDngb2AfwKwPVXSdsCNkra1vaK+Q0kzgZkAbRMn0bFi9nB+v4iI0em4635/ufjEvUdwIBExFAb0FLztMyTtCGwPrF+XfuFwDWyU2gO42PZqYImkm+vy9pT0GaotC38MPMyrAegV5XMe0FGudwPOALD9qKSfAdsCC+s7tD0LmAUwYYspA3tiLCIiImIUG1AAKunzwAyqAPQHwN8AdwLjLQCF8i7UemWv7NlAzfaTkjqpC9R59Z2pq3n1N8+L/CMiImJcGuiL6A8C/hL4he3DgbcCE4ZtVKPX7cAhktokbQF079vsDjafk7Qx1e81kLYOBZC0LbA18NgQjzciIiJi1Bnoi+hfsv2KpFWSJgLPAn8+jOMara4E3g0sAn4C3AZg+9eSzi3pi4G5A2jrbOCbkhYBq4DDbPd5utTUye10Ze9TREREtLiBBqBdkjYFzqXax7gcuH/YRjXK2N64fBr4RC9l/g34twbpM+qun6PsAS0PGx025IONiIiIGOUG+hDSP5bLb0q6Hphoe2FfdSIiIiIiGhnwSUjd17YX2144nk5CioiIiIihk5OQIiIiIqKp+luC/xhwNFWwOY8qADXwG+DM4R1aRERERIxFfS7B2z7N9huBLwHTyvVs4HHgniaMLyIiIiLGmAG/B9T2C5J2A94DzAHOGbZRRURERMSYNdDXMK0un3sD37T9/XLaTzTTkvnQ2T7So4iIGB86l430CCLGrIHOgD4l6d+BDwA/kDRhEHVbliRLuqjufl1Jv5R0bT/1apJOH/4RRkRERLSegQaRHwBuAP6n7V8Dfwz8y7CNavR4EdhR0gbl/j3AU/1Vst1l+6hhHVlEREREixpQAGr7t7avsP3Tcv+07RuHd2ijxn9SbT0A+CBwcXeGpJ0l3S1pfvl8U0mf0T1LKqlT0vmSbpX0uKSj6ur/naT7JS2Q9O+S2pr4vSIiIiJGxED3gI5nlwDHl4DyLcD5wO4l71FgD9urJO0FfBk4sEEb2wF7ApsAj0k6B/gL4GBgV9svSzobOBS4sL6ipJnATIC2iZPoWDF7qL9fRMQaW3zi3v0XiojoIQFoP8qpTx1Us58/6JHdDlwgaQrV+1HX66WZ62yvBFZKehbYHPhLYCdgriSADYBnG/Q/C5gFMGGLKV7b7xMREREx0hKADszVwNeBGcDr69K/ANxie/8SpN7aS/2VdderqX53ARfY/tchHmtERETEqJYAdGDOB5bZXiRpRl16O68+lHTYINu8Cfi+pFNsPyvpj4FNbP+stwpTJ7fTleWuiIiIaHFj/lVKQ8H2z22f1iDrJOArku4CBvUAke1HgH8DbpS0EPghsMVaDzYiIiJilJOdbYWtolaruaura6SHEREREdEvSfNs1xrlZQY0IiIiIpoqAWhERERENFUC0IiIiIhoqgSgEREREdFUCUAjIiIioqnyHtBWsmQ+dLaP9CgiIkZG57KRHkFEDJExOwMq6RRJR9fd3yDpvLr7kyV9apj6Pk/S9sPRdkRERESrG7MBKHA3sAuApHWAzYAd6vJ3Ae4ajo5t/0N50XxERERE9DCWA9C7KAEoVeD5EPAbSa+TNAF4M3CqpGndFSTdJektkv5Y0lWSFkq6V9JbSn6npAsk3ShpsaQDJJ0kaZGk6yWtV8rdKqlWrpdL+pKkB0tbm5f0bcr9XEknSFrevJ8mIiIiYuSM2T2gtpdIWiVpa6pA9B5gMjAdWAYsBP6D6gz3oyVtC0ywvVDSGcB82/tJejdwIdAdqG4D7AlsX9o80PZnJF0J7A1c1WMoGwH32v6spJOAjwJfBE4DTrN9saQje/sekmYCMwHaJk6iY8XstfthIiJa1XHX9Zm9+MS9mzSQiFhbY3kGFF6dBe0OQO+pu78buAzYp8xcHgHMKfV2Ay4CsH0z8HpJ3U///Kftl4FFVOe/X1/SFwEdDcbwO+Dacj2vrsz00j/Ad3r7ArZn2a7ZrrVtmAeQIiIiovWN2RnQonsf6FSqJfgngU8DLwDn2/6tpB8C+wIfALrPK1WDtlw+VwLYfkXSy7a701+h8e9ZX2Z1L2UiIiIixo3xMAO6D7DU9mrbS4FNqWYf7yllzgNOB+aWfIDbgUMBJM0AnrP9whCP7V7gwHJ9yBC3HRERETFqjfXZuEVUT79/p0faxrafA7A9T9ILQP3myk5gtqSFwG+BDw/D2I4GviXp08B1VPtS+zR1cjtd2eMUERERLU6vrg6PT5K2BG4FtrP9ShP73RB4ybYlHQJ80Pa+fdWp1Wru6upqzgAjIiIi1oKkebZrjfLG+gxonyR9CPgS8KlmBp/FTsCZkgT8muohqIiIiIgxb1wHoLYvpHrF0kj0fQfw1pHoOyIiImIkjfWHkCIiIiJilEkAGhERERFNlQA0IiIiIpoqAWhERERENNW4fgip5SyZD505jjMiYtTr7PfVzhHj2qiaAZW0WtICSQ9Juqy8K3OkxnKYpDOHuM0tJX1vKNuMiIiIaDWjKgClejH7NNs7Ar8DjhzpAQ0l20tsHzTS44iIiIgYSaMtAK13B/AXkjaSdL6kuZLmS9oXfj9DeYWk6yX9VNJJ3RUlLZf0JUkPSrpX0uYl/f1ldvVBSbeXtDskTaure5ekt9Tdt0taLGmdcr+hpCclrSfpo2VcD0q6vHvGVtIcSadLulvS45IOKukdkh6qu75D0gPlb5dh/0UjIiIiRoFRuQdU0rrA3wDXA58FbrZ9hKRNgfsl/agUnQa8DVgJPCbpDNtPAhsB99r+bAlMPwp8ETge+GvbT5W2AM4DDgOOlrQtMMH2QklvB7C9TNKDwLuAW4D3ATfYflnSFbbPLWP+IvAR4IzS7hbAbsB2wNVAz6X3Z4H32F4haQpwMfCa46okzQRmArRNnETHitk9i0REjEqLT9x7pIcQEaPUaJsB3UDSAqAL+G/gP4C/Ao4r6bcC6wNbl/I32V5mewXwCPCGkv474NpyPQ/oKNd3AXMkfRRoK2mXAftIWo/qOMw5DcZ1KXBwuT6k3APsWGYxFwGHAjvU1bnK9iu2HwE2b9DmesC5pe5lwPaNfhDbs2zXbNfaNswDSBEREdH6RtsM6Eu2p9UnlLPSD7T9WI/0/0E189ltNa9+n5dtu2e67SNLvb2BBZKm2X5e0g+BfYEP0GAWkmoG8yuS/pjqDPebS/ocYD/bD0o6DJhRV6d+bGrQ5jHAM1THca4DrGhQJiIiImLMGW0BaCM3AJ+U9EnblvQ22/PXpCFJ29i+D7hP0vuArYDnqZbhrwHusL20Zz3byyXdD5wGXGt7dcnaBHi6zJ4eCjw1iOG0Az+3/YqkD/PqjGyvpk5upytLWhEREdHiWiEA/QJwKrCwzIYuBvZZw7a+VvZbCrgJeBDA9jxJLwB9bbC8lGqpfEZd2ueA+4CfAYuoAtKBOhu4XNL7qfaWvjiIuhEREREtS6+uVI9fkrak2l+6ne1XRng4varVau7q6hrpYURERET0S9I82422No66h5CaTtKHqGYxPzuag8+IiIiIsaIVluCHle0LgQtHehwRERER48W4nwGNiIiIiOZKABoRERERTZUANCIiIiKaatzvAW0pS+ZDZ05DiogYMp3LRnoEEeNSZkABSX8m6fuSfirpvySdJumPRnpcEREREWPRuA9Ay8vtr6A6u30KsC2wMfClHuUyWxwRERExBMZ9AAq8G1hhezZAOWbzGOAISf8o6TJJ1wA3StpI0vmS5kqaL2lfAEkbSvqupIWSLpV0n6RayfugpEWSHpL01e5OJS2X9CVJD0q6V9Lmzf/qEREREc2XWT3YAZhXn2D7BUn/TfX7TAfeYnuppC8DN9s+QtKmwP2SfgR8HPiV7bdI2hFYAL8/YemrwE7Ar6iC2P1sXwVsBNxr+7OSTgI+Cnyx5+AkzQRmArRNnETHir5OC42IiEYWn7j3SA8hIupkBrQ6F77ReaTd6T+0vbSk/RVwnKQFVEd3rg9sDewGXAJg+yFgYSn/DuBW27+0vQr4NrBHyfsdcG25ngd0NBqc7Vm2a7ZrbRvmAaSIiIhofZkBhYeBA+sTJE0EtgJWAy/WZwEH2n6sR3n10nZv6QAv2+4OfFeT/0VERESME5kBhZuADcuZ8EhqA04G5gC/7VH2BuCT3QGnpLeV9DuBD5S07YGpJf0+4F2SNivtfhC4bfi+SkRERMToN+5n3Wxb0v7A2ZI+RxWU/wD4P1QBY70vAKcCC0sQuhjYBzgbuEDSQmA+1RL8MttPS/pX4Baq2dAf2P7+mo516uR2urKPKSIiIlqcXl0FjjVVZjfXs71C0jZUs6rb2v7dUPZTq9Xc1dU1lE1GREREDAtJ82zXGuWN+xnQIbIhcIuk9ahmOj8+1MFnRERExFiRAHQI2P4N0DDCj4iIiIg/lIeQIiIiIqKpEoBGRERERFMlAI2IiIiIpkoAGhERERFNlYeQWsmS+dCZ4zgjIgLoXDbSI4hYY+N2BlTSrZL+ukfa0ZIel3TcINvaUtL3BlDuB5I2HexYIyIiIsaS8TwDejFwCNXxmt0OAT5s+46ehSWta3tVo4ZsLwEO6q9D2+9dw7FGREREjBnjdgYU+B6wj6QJAJI6gC2Bv5B0ZkmbI+kbkm4BvippG0n3Spor6QRJy7vrSnqoXB8m6QpJ10v6qaSTujuUtFjSZuX6KknzJD0saWYzv3hERETESBq3M6C2n5d0P/A/ge9TzX5eCvQ8m3RbYC/bqyVdC5xm+2JJR/bR/DTgbcBK4DFJZ9h+skeZI2wvlbQBMFfS5baf79lQCU5nArRNnETHitlr8G0jIobW4hP3HukhREQLG88zoPDqMjzl8+IGZS6zvbpcTwcuK9ff6aPdm2wvs70CeAR4Q4MyR0l6ELgX2AqY0qgh27Ns12zX2jbMA0gRERHR+sZ7AHoV8JeS3g5sYPuBBmVeXIN2V9Zdr6bHTLOkGcBewHTbbwXmA+uvQT8RERERLWdcB6C2lwO3AufTePazp3uBA8v1IX0V7Ec78Cvbv5W0HfDOtWgrIiIioqWM2z2gdS4GrmBgAeXRwLckfRq4DljTl7BdDxwpaSHwGFVg26+pk9vpyr6riIiIaHGyez5zE72RtCHwkm1LOgT4oO19m9V/rVZzV1dXs7qLiIiIWGOS5tmuNcrLDOjg7AScKUnAr4EjRng8ERERES0nAegglBfUv3WkxxERERHRysb1Q0gRERER0XwJQCMiIiKiqRKARkRERERTZQ9oK1kyHzpzGlJExJjSuaZv9ItoXS0xAyrJkk6uuz9WUucQtt9R+vhCXdpmkl6WdOYatnmCpL2GaowRERERY0VLBKBUR1seIGmzYezjcWCfuvv3Aw+vaWO2j7f9o7UeVURERMQY0yoB6CpgFnBMzwxJkyRdLmlu+du1pC+StKkqz0v6UEm/qJeZyZeAH0vqfmHqwcB3B9DP9+va/pikb5frOZIOKtfvkHS3pAcl3S9pE0nrS5pdxjlf0p5D9WNFREREjGattAf0LGChpJN6pJ8GnGL7TklbAzcAbwbuAnYFfkY1u7k7cCHVuesf76WPS4BDJP0CWA0sAbbsp5+ZwF2SngA+TY9z3SX9EXApcLDtuZImUgW7/wxge2o5D/5GSdvaXtGj/szSB20TJ9GxYvbAfq2IiGgJi0d6ABEjoGUCUNsvSLoQOIoqgOu2F7B9dTgRABMlbQLcAexBFYCeA8yUNBlYant5L91cD3wBeIYqaKzXsB/bz0g6HrgF2N/20h713gQ8bXtu9/cAkLQbcEZJe1TSz4BtgYU9vvcsqtlfJmwxJeemRkRERMtrmQC0OBV4AKifBlwHmG67PihF0u3APwFbA58F9gcOogpMG7L9O0nzqGYydwDe118/xVTgeV6dLf2DoQCNAkc1SIuIiIgY81plDygAZXbxu8BH6pJvBD7RfSNpWin7JLAZMMX248CdwLH0EYAWJwP/2/bzPdIb9iNpZ+BvgLcBx0p6Y496jwJbSnpHKb+JpHWB24FDS9q2VIHyY/2MLSIiIqLltdoMKFQB4ifq7o8CzpK0kOr73A4cWfLuA9rK9R3AV6gC0V7ZfpjGT7+/ph9J/wycCxxue4mkTwPnS3p3XXu/k3QwcIakDai2D+wFnA18U9IiqoesDrO9sq+xTZ3cTteJe/dVJCIiImLUk51tha2iVqu5q6trpIcRERER0S9J82zXGuW11BJ8RERERLS+BKARERER0VQJQCMiIiKiqRKARkRERERTJQCNiIiIiKZKABoRERERTdWK7wEdv5bMh872kR5FRES0os5lIz2CiN9rmRlQSadIOrru/gZJ59XdnyzpU4Nor1PSsb3k3b0W45whaZc1rR8REREx1rVMAArcDewCIGkdqmM2d6jL3wW4ayg6sr02AeSMMpaIiIiIaKCVAtC7eDWw2wF4CPiNpNdJmgC8GfhrSXMlPSRpliQBSDpK0iOSFkq6pK7N7SXdKulxSUd1J0paXj5nlPzvSXpU0rfr2nxvSbtT0umSrpXUQXUM6DGSFkjaXdIbJN1U+r5J0tal/pxS7+7S/0HD+utFREREjBItswe0nLW+qgRwuwD3AJOB6cAyYCFwpu0TACRdBOwDXAMcB7zR9kpJm9Y1ux2wJ7AJ8Jikc2y/3KPrt1EFvEuoguBdJXUB/w7sYfsJSReXMS6W9E1gue2vl3FcA1xo+wJJRwCnA/uVtrcAdivjuBr4Xs/vLWkmMBOgbeIkOlbMXpOfLyIiRpHFJ+490kOIGFGtNAMKr86Cdgeg99Td3w3sKek+SYuAd/PqEv1C4NuS/g5YVdfedbZX2n4OeBbYvEGf99v+ue1XgAVAB1XA+LjtJ0qZi/sY83TgO+X6IqqAs9tVtl+x/UgvfWN7lu2a7VrbhnkAKSIiIlpfqwWg3ftAp1Itwd9LFeB17/88GzjI9lTgXGD9Um9v4CxgJ2CepO6Z35V1ba+m8YxwozJai+/gXtpemzYjIiIiWkarBaB3US2rL7W92vZSYFOqIPSeUuY5SRsDB8HvH1jayvYtwGdK+Y3XchyPAn9e9nwCHFyX9xuqJf1udwOHlOtDgTvXsu+IiIiIltYye0CLRVRPv3+nR9rGtp+TdG65XwzMLfltwLcktVPNMp5i+9flWaI1YvslSf8IXC/pOeD+uuxrgO9J2hf4JHAUcL6kfwF+CRy+pv1OndxOV/YNRURERIuT7f5LxWtI2tj28vJU/FnAT22fMpx91mo1d3V1DWcXEREREUNC0jzbtUZ5rbYEP5p8VNIC4GGgneqp+IiIiIjoR6stwY8aZbZzWGc8IyIiIsaizIBGRERERFMlAI2IiIiIpkoAGhERERFNlT2grWTJfOjMaUgRETHOdS4b6RHEWhrTM6CSXM6E775fV9IvJV07RO3PGMK2DpN05lC0FRERETGajekAFHgR2FHSBuX+PcBTg2mg7tjOiIiIiBgCYz0ABfhPqrPgAT4IXNydIWlnSXdLml8+31TSD5N0maRrgBslXVRONuqu921Jf1vfST9tXSHpekk/lXRSXZ3DJf1E0m3ArsP2C0RERESMIuNhdu8S4PiyVP4W4Hxg95L3KLCH7VWS9gK+DBxY8qYDb7G9VNK7gGOA75cjPXcBPgzsVtdPX21NA94GrAQek3QGsAr4v8BOwDLgFmB+z8FLmgnMBGibOImOFbPX9veIiBiXFuco44hRY8wHoLYXSuqgmv38QY/sduACSVMAA+vV5f3Q9tLSxm2SzpL0J8ABwOUl0BxoWzfZXgYg6RHgDVRn2t9q+5cl/VJg2wbjnwXMApiwxZScmxoREREtbzwswQNcDXyduuX34gvALbZ3BN4HrF+X92KPshcBhwKHA42mIftqa2Xd9WpeDfwTUEZERMS4M14C0POBE2wv6pHezqsPJR3WTxtzgKMBbD/cIH8wbQHcB8yQ9HpJ6wHvH0CdiIiIiJY35pfgAWz/HDitQdZJVMvmnwJu7qeNZyT9GLiqlyIDbqu097SkTuAe4GngAaCtrzpTJ7fTlT1MERER0eJkZxV4ICRtCCwC3t69n7PZarWau7q6RqLriIiIiEGRNM92rVHeeFmCXyvlqfZHgTNGKviMiIiIGCvGxRL82rL9I2DrkR5HRERExFiQGdCIiIiIaKoEoBERERHRVAlAIyIiIqKpEoBGRERERFPlIaRWsmQ+dLaP9CgiIiLWXmdeKjOeDdsMqCRLOrnu/tjy4vWhar9D0kM90jolHTtUfQxiLL32K+nuZo8nIiIiYjQbziX4lcABkjYbxj5GPdu79EyT1OeJRxERERFj2XAGoKuAWcAxPTMkTZJ0uaS55W/Xkr5I0qaqPC/pQyX9ovIy+AGT9NHS9oOlrw1L+hxJ50i6RdLjkt4l6XxJP5Y0p67+ckknS3pA0k2SJpX0oyQ9ImmhpEvqutxe0q2lzaPq2ymfM0qf3wEWSWqT9LUyxoWSPjaY7xcRERHRqoZ7D+hZwEJJJ/VIPw04xfadkrYGbgDeDNwF7Ar8DHgc2B24EHgn8PEG7W8jaUHd/Z8CXy/XV9g+F0DSF4GPAGeUvNcB7wb+Frim9PkPwFxJ02wvADYCHrD9aUnHA58HPgEcB7zR9kpJm9b1vR2wJ7AJ8Jikc2y/3GO8OwM72n5C0kxgme13SJoA3CXpRttP1Fco5WYCtE2cRMeK2Q1+hoiIaFWLT9x7pIcQ0XTDGoDafkHShcBRwEt1WXtRzRh230+UtAlwB7AHVQB6DjBT0mRgqe3lDbr4L9vTum967DHdsQSemwIbUwW53a6xbUmLgGdsLyr1HwY6gAXAK8Clpfy3gCvK9ULg25KuAq6qa/M62yuBlZKeBTYHft5jvPfXBZh/BbxF0kHlvh2YAvxBAGp7FtVMMhO2mOIGv0FERERES2nGU/CnAg8A9VN36wDTbdcHpUi6HfgnqmMvPwvsDxxEFZgO1hxgP9sPSjoMmFGXt7J8vlJ33X3f22/SHfztTRUk/y3wOUk79GgTYHUv7bxYdy3gk7ZvaFAuIiIiYswa9veA2l4KfJdqCbzbjVTL2QBImlbKPglsBkyx/ThwJ3AsaxaAbgI8LWk94NA1qL8OVfAL8L+AOyWtA2xl+xbgM7w6u7ombgA+XsaHpG0lbbSGbUVERES0jGa9B/Rk6gJOqiX5syQtLGO4HTiy5N0HdD8lfgfwFapAdLA+V9r6GbCIKiAdjBeBHSTNA5YBB5dxfUtSO9UM5im2f123lWAwzqNa7n9AVQO/BPbrq8LUye10Za9QREREtDjZ2VbYiKTlttd0dnNY1Go1d3V1jfQwIiIiIvolaZ7tWqO8HMUZEREREU2VALQXo232MyIiImKsSAAaEREREU2VADQiIiIimioBaEREREQ0VbNewxRDYcl86Gwf6VFERESMLp3LRnoEMUijZgZU0imSjq67v0HSeXX3J0v61CDbnCFplwbpHZJ+Xl4sX5++QNLOks6TtP0g+zpS0of6KVOTdPpg2o2IiIgYa0bTDOjdwPuBU0tguBkwsS5/F+DoRhX7MANYXtr+PduLJT0J7A7cBiBpO2AT2/cD9zdqTFKb7dWN8mx/s7/B2O4C8iLPiIiIGNdGzQwocBdVkAmwA/AQ8BtJr5M0AXgzMF/STpJukzSvzJJuASDpKEmPSFoo6RJJHVSnKx1TZjZ379HfxcAhdfeHlDQk3SqpVq6XSzpB0n3AdEkfkfSTUuZcSWeWcp2Sjq2r/1VJ95eyu5f0GZKuLdc7S7pb0vzy+aYh/TUjIiIiRqlRMwNqe4mkVZK2pgpE7wEmA9OpjsJcCBg4A9jX9i8lHQx8CTgCOA54o+2VkjYtR2R+E1hu++sNuvwuVUD7SdurqI7afH+DchsBD9k+XtKWwLeAtwO/AW4GHuzlK61re2dJ7wU+D+zVI/9RYA/bqyTtBXwZOLBnI5JmAjMB2iZOomPF7F66i4iIZlqco5Ej1tioCUCL7lnQXYBvUAWgu1AFoHcDbwJ2BH5Yzl9vA54udRcC35Z0FXBVfx3Z/oWkh4G/lPQM8LLthxoUXQ1cXq53Bm6zvRRA0mXAtr10cUX5nEd15ntP7cAFkqZQBdbr9TLOWcAsgAlbTMm5qREREdHyRlsAejdVwDmVagn+SeDTwAvA+YCAh21Pb1B3b2AP4G+Bz0naYQD9dS/DP1OuG1lRt+9TA/weACvL52oa/85fAG6xvX/ZLnDrINqOiIiIaFmjaQ8oVDOg+wBLba8uM42bUi3D3wM8BkySNB1A0nqSdigPLW1l+xbgM6XOxlTL5Jv00d/lwHuplt8vGcD47gfH9NX+AAAYB0lEQVTeVfalrkuDJfNBaAeeKteHrUU7ERERES1ltM2ALqJ6+v07PdI2tv0cgKSDgNMltVON/1TgJ8C3SpqAU8oe0GuA70naF/ik7TvqOytl7gU2t/1Ef4Oz/ZSkLwP3AUuAR6i2B6yJk6iW4D9FtZe0X1Mnt9OVPUcRERHR4mRnW+FgSNrY9vIyA3olcL7tK5vRd61Wc1dX3uIUERERo5+kebZrjfJG2xJ8K+iUtIBqj+oTDOCBp4iIiIh41Whbgh/1bB870mOIiIiIaGWZAY2IiIiIpkoAGhERERFNlQA0IiIiIpoqAWhERERENFUeQmolS+ZDZ/tIjyIiIqJ1dK7p67pjOI3rGVBJlnRy3f2xkjqHuc/FkjYbzj4iIiIiRrNxHYBSndd+QALCiIiIiOYZ7wHoKmAWcEzPDEmTJF0uaW7521XSOmUGc9O6cv9P0uaS3ifpPknzJf1I0uYl//WSbizp/051VGh33askzZP0sKSZTfi+ERERESMue0DhLGChpJN6pJ9Gdab8nZK2Bm6w/WZJ3wf2B2ZL+h/AYtvPSLoTeKdtS/oH4DPAp4HPA3faPkHS3kB9oHmE7aWSNgDmSrrc9vP1gyiB6UyAtomT6Fgxe+h/gYiIiFFu8Yl7j/QQYgiN+wDU9guSLgSOAl6qy9oL2F76/YTlREmbAJcCxwOzgUPKPcCfAZdK2gL4I6pjOgH2AA4ofV0n6Vd1fRwlaf9yvRUwBfiDANT2LKpZWiZsMcVr920jIiIiRt54X4LvdirwEWCjurR1gOm2p5W/ybZ/A9wD/IWkScB+wBWl/BnAmbanAh8D1q9r6zWBo6QZVEHudNtvBeb3qBMRERExJiUABWwvBb5LFYR2uxH4RPeNpGmlrIErgW8AP65bMm8HnirXH65r53bg0NLG3wCvqyv/K9u/lbQd8M6h/E4RERERo9W4X4KvczJ1ASfVkvxZkhZS/U63A0eWvEuBucBhdeU7gcskPQXcC7yxpP9f4GJJDwC3Af9d0q8HjiztP1bq9Gnq5Ha6sgcmIiIiWpyqCb1oBbVazV1dXSM9jIiIiIh+SZpnu9YoL0vwEREREdFUCUAjIiIioqkSgEZEREREUyUAjYiIiIimSgAaEREREU2V1zC1kiXzobN9pEcRERHRejqXjfQIos6YnQGV9KeSLpH0X5IekfQDSdv2UX55+dxS0vfq0i+WtFDSMUMwppqk09e2nYiIiIhWNiZnQFUd4H4lcIHtQ0raNGBz4Cd91bW9BDio1PlTYBfbbxhE3+vaXtVL211AXuQZERER49pYnQHdE3jZ9je7E2wvAOZLuknSA5IWSdq3Z0VJHZIeKrc3An8iaYGk3SVNk3RvmRG9UtLrSp1bJX1Z0m3AP5f7r0q6X9JPJO1eys2QdG253lnS3ZLml883DfNvEhERETEqjMkZUGBHYF6D9BXA/rZfkLQZcK+kq937cVB/C1xrexpAOTbzk7Zvk3QC8Hng6FJ2U9vvKuXeB6xre2dJ7y3l9urR9qPAHrZXSdoL+DJwYM8BSJoJzARomziJjhWzB/obREREtLzFOYJ6TBqrAWhvBHxZ0h7AK8BkqmX5X/RbUWqnCjJvK0kXAJfVFbm0R5Uryuc8oKNBk+3ABZKmAAbWa9Sv7VnALIAJW0zJuakRERHR8sbqEvzDwE4N0g8FJgE7lVnNZ4D1h6jPF3vcryyfq2kc6H8BuMX2jsD7hnAcEREREaPaWA1AbwYmSPpod4KkdwBvAJ61/bKkPcv9gNheBvyqez8n8PfAbX1U6U878FS5Pmwt2omIiIhoKWNyCd62Je0PnCrpOKq9n4uBTuB0SV3AAqp9mIPxYeCbkjYEHgcOX4thnkS1BP8pqoC5X1Mnt9OVvTARERHR4tT78zcx2tRqNXd15S1OERERMfpJmme71ihvrC7BR0RERMQolQA0IiIiIpoqAWhERERENFUC0IiIiIhoqgSgEREREdFUCUAjIiIioqnG5HtAx6wl86GzfaRHERERMfp0LhvpEcQgtOwMqCRLOrnu/lhJnUPYfoekh4aqvYiIiIiotGwASnXW+gGSNhvpgQBIahvpMURERES0glYOQFcBs4BjemZImiTpcklzy9+uJX2RpE1VeV7Sh0r6RZL26q0jSW2SvlbaWijpYyV9hqRbJH0HWCRpI0nXSXpQ0kOSDi7ldpJ0m6R5km6QtIWkbSQ9UNfHFEnzhvQXioiIiBiFWn0P6FnAQkkn9Ug/DTjF9p2StgZuAN4M3AXsCvyM6iz33YELgXcCH++jn48Ay2y/Q9IE4C5JN5a8nYEdbT8h6UBgie29ASS1S1oPOAPY1/YvS1D6JdtHSFomaZrtBVTnys/p2bGkmcBMgLaJk+hYMXtwv1BERMQ4sHikBxCD0tIBqO0XJF0IHAW8VJe1F7C9pO77iZI2Ae4A9qAKQM8BZkqaDCy1vbyPrv4KeIukg8p9OzAF+B1wv+0nSvoi4OuSvgpca/sOSTsCOwI/LONpA54u5c8DDpf0KeBgqmC253ecRTXTy4QtpngAP0tERETEqNbSAWhxKvAAUD81uA4w3XZ9UIqk24F/ArYGPgvsDxxEFZj2RcAnbd/Qo70ZwIvd97Z/Imkn4L3AV8os6ZXAw7anN2j3cuDzwM3APNvP9zOOiIiIiJbXyntAAbC9FPgu1TJ5txuBT3TfSJpWyj4JbAZMsf04cCdwLP0HoDcAHy/L6UjaVtJGPQtJ2hL4re1vAV8H3g48BkySNL2UWU/SDmU8K0rb5/CHAXRERETEmDUWZkABTqYu4KRakj9L0kKq73g7cGTJu49qGRyqwPMrVIFoT+tSPWkP1VJ5B/CAqnX0XwL7NagzFfiapFeAl4GP2/5dWbo/XVJ7afdU4OFS59vAAVRBc5+mTm6n68S9+ysWERERMarJzrbCRiTtCxxq+wPD3M+xQLvtz/VXtlaruauraziHExERETEkJM2zXWuUN1ZmQIeUpBOAfYHDhrmfK4FtgHcPZz8RERERo0kC0AZsHw8c34R+9h/uPiIiIiJGm5Z/CCkiIiIiWksC0IiIiIhoqgSgEREREdFUCUAjIiIioqnyEBIgaX/gCuDNth/tpczdtndp7sh6WDIfOttHdAgRERHjQueykR7BmJYZ0MoHqV5Gf0jPDEltACMefEZERESMEeM+AJW0MbAr1VGeh5S0GZJukfQdYFFJW14+T5C0oPw9JWl2Sf+UpIfK39ElrUPSjyWdK+lhSTdK2qDkfVTSXEkPSrpc0obN//YRERERzTfuT0KS9HfAnrY/IuluqiM9JwLXATvafqKUW25747p67VRHeR5ekuYA7wREddzn3wG/Av4fULO9QNJ3gattf0vS620/X9r6IvCM7TMajG8mMBOgbeKknf7s4zkyPiIiYk0sznHWTdXXSUjjfgaUavn9knJ9SbkHuL87+OypnAf/beAU2/OA3YArbb9oeznVftLdS/EnbC8o1/OozpQH2FHSHZIWAYcCOzTqy/Ys2zXbtbYNs/8zIiIiWt+4fghJ0uupjsHcUZKBNsDAD4AX+6jaCfzcdvd0pPoou7LuejWwQbmeA+xn+0FJhwEzBjn8iIiIiJY03mdADwIutP0G2x22twKeoJrRbEjSPsB7gKPqkm8H9pO0oaSNgP2pluf7sgnwtKT1qGZAIyIiIsaFcT0DSrXcfmKPtMuBjwP/1UudTwNbAvdXK/Fcbft4SXOA+0uZ82zPl9TRR9+fo9or+jOqB5026W+wUye305X9KxEREdHixv1DSK2kVqu5q6trpIcRERER0a88hBQRERERo0YC0IiIiIhoqgSgEREREdFUCUAjIiIioqkSgEZEREREUyUAjYiIiIimGu/vAW0tS+ZDZ47jjIiI/9/evQfbVZZ3HP/+ehBDDAQC1EIQYkVFhBLwgHIdAswUhBZ0yoCDLdBLytSpRkspagcjMzLM0FrtVIrxAtTSaKFcO4rcCWi5nJBgghCVi0K5BVHuIRie/rFXhs3pSXLIZe+sc76ff/Ze73rXWs/Zz5xznnnfd+2lMWH2M/2OoG/GzQhokpuS/P6wtllJzu1XTF0xTOxnDJIkSb00bgpQYC5w/LC245v21UoysEEi6pgFWIBKkqRxYzwVoJcARyV5M0DzmMztgYlJ/ifJXUkuTjKp2f9QkjOS3Aoc22yf1fQdSrJXku8nuT/JKc0xSXJOksVJFiU5rmk/uBmBvSTJfUkuavp+vInhxiQ39v4jkSRJ6r1xswa0qn6Z5A7gcOAKOqOf1wOfBQ6rqheS/B3wKeDM5rBlVXUAQJKzgYerat8k/wRcAOwPTADuAc4DPgxMB/YAtgHuTDKvOdeewHuBR4EfAPtX1T8n+RQwo6qeGinuJDOBmQADW2zLtGXnr6+PRJIkrcJDZx/Z7xDGtPE0Agqvn4Y/HngQ2BX4QZKFwInATl39vzPs+Cub10XA7VX1XFUtBZYl2RI4AJhbVSuq6gngZmDv5pg7quqRqnoVWAhMG03AVTWnqgaranBgojcgSZKk9hs3I6CNy4EvJtkL2AxYAFxbVR9ZRf8Xhm2/3Ly+2vV+5fYmQFZz7e7+Kxh/n70kSRIwzkZAq+p54Cbgm3RGQ28D9k+yM0CSiUnetQ6XmAccl2QgybbAQcAdazjmOWDzdbimJElSq4zHUbi5wKXA8VW1NMlJwNyVNycBfw/8ZC3PfRmwL3A3UMBpVfV4kl1Wc8wc4HtJHquqGas7+e5TJzPkmhRJktRyqap+x6BRGhwcrKGhoX6HIUmStEZJ5lfV4Ej7xtUUvCRJkvrPAlSSJEk9ZQEqSZKknrIAlSRJUk9ZgEqSJKmnLEAlSZLUU+Pxe0Db69EFMNvHcUqSJGD2M/2OYK2NixHQJL+T5NtJ7k/y4yTfXdUTj5JMS7J4A8UxO8mpG+LckiRJbTHmC9AkofOEopuq6h1VtSvwGeCt6+n8jiJLkiS9AWO+AAVmAK9U1XkrG6pqIXBrknOSLE6yKMlxww9MMiHJ+c3+BUlmNO0nJbk4yVXANUkmJbk+yV1N36O7zvHZJEuSXAe8u6t9epLbkvwoyWVJttqQH4IkSdLGYjyM3u0GzB+h/cPAdGAPYBvgziTzhvX5GEBV7d48z/2arqn7fYHfq6qnm1HQD1XVs0m2AW5LciWwF3A8sCedz/qurlj+Dfjrqro5yZnA54BZw4NMMhOYCTCwxbZMW3b+Wn0IkiRpbHmo3wGsg/EwAroqBwBzq2pFVT0B3AzsPUKfbwFU1X3Az4GVBei1VfV08z7AWUl+BFwHTKUzxX8gcFlVvVhVzwJXAiSZDGxZVTc3x18IHDRSkFU1p6oGq2pwYKI3IEmSpPYbDwXoPcD7RmjPKI5dXZ8Xut6fAGwLvK+qpgNPABOafTWaICVJksaL8TAFfwOd0cm/qKqvASTZG/gVcFySC4EpdEYg/5bXCkeAeXSKyxuaqfcdgSV0pta7TQaerKpXmnWiO3Udf0GSs+l81n8AfLWqnknyqyQHVtUtwB/TGYFdrd2nTmbo7CPX4iOQJEnaeIz5ArSqKsmHgC8lOR1YRmfZxCxgEnA3nVHK06rq8STTug4/FzgvySLgN8BJVfVy58b617kIuCrJELAQuK+59l1JvtO0/Ry4peuYE5tzTwQeAE5ebz+0JEnSRixVzhC3xeDgYA0NDfU7DEmSpDVKMr+qBkfaNx7WgEqSJGkjYgEqSZKknrIAlSRJUk9ZgEqSJKmnLEAlSZLUU2P+a5jGlEcXwGyfhiRJktbB7Gf6HYEjoGsjybQki4e1zU5yapKTkmw/inPclGTEryaQJEkayyxA17+TgDUWoJIkSeOVBej6NwhclGRhks2SnJHkziSLk8zJ6x+jdGySO5L8JMmB/QpYkiSpl1wDuv4NAadW1RBAkn+pqjOb998CjgKuavpuUlX7JPkg8DngsOEnSzITmAkwsMW2TFt2fg9+BEmS1E8PnX1kv0PYoBwBXTuren7pSO0zktzePE/+EOC9XfsubV7nA9NGPGHVnKoarKrBgYnegCRJktrPEdC180tgq2FtU4AHuxuSTADOBQar6uEks4EJXV1ebl5XYC4kSdI44QjoWqiq54HHkhwKkGQKcDhwK/AcsHnTdWWx+VSSScAf9TpWSZKkjY2jbmvvT4CvJPnHZvvzVXV/kguA85K8BOwLfA1YBDwE3LkuF9x96mSGxviaEEmSNPalalXLGbWxGRwcrKGhoX6HIUmStEZJ5lfViN957hS8JEmSesoCVJIkST3lFHyLJHkOWNLvOLTOtgGe6ncQWi/M5dhgHscG87jx2amqth1phzchtcuSVa2lUHskGTKPY4O5HBvM49hgHtvFKXhJkiT1lAWoJEmSesoCtF3m9DsArRfmcewwl2ODeRwbzGOLeBOSJEmSesoRUEmSJPWUBagkSZJ6ygK0JZIcnmRJkp8lOb3f8Wh0krwtyY1J7k1yT5JPNO1Tklyb5KfN61b9jlVrlmQgyYIk/91sm8eWSbJlkkuS3Nf8Xu5rHtsnySebv6mLk8xNMsE8tosFaAskGQC+AhwB7Ap8JMmu/Y1Ko/Qb4G+q6j3AB4CPNbk7Hbi+qt4JXN9sa+P3CeDerm3z2D5fBq6uql2APejk0zy2SJKpwMeBwaraDRgAjsc8tooFaDvsA/ysqh6oquXAt4Gj+xyTRqGqHququ5r3z9H5ZzeVTv4ubLpdCBzTnwg1Wkl2AI4Evt7VbB5bJMkWwEHANwCqanlV/Rrz2EabAJsl2QSYCDyKeWwVC9B2mAo83LX9SNOmFkkyDdgTuB14a1U9Bp0iFfjt/kWmUfoScBrwalebeWyX3wWWAuc3Sym+nuQtmMdWqar/Bf4B+AXwGPBMVV2DeWwVC9B2yAhtfn9WiySZBPwXMKuqnu13PHpjkhwFPFlV8/sdi9bJJsBewL9W1Z7ACzhN2zrN2s6jgbcD2wNvSfLR/kalN8oCtB0eAd7Wtb0DnekGtUCSN9EpPi+qqkub5ieSbNfs3w54sl/xaVT2B/4wyUN0lsAckuTfMY9t8wjwSFXd3mxfQqcgNY/tchjwYFUtrapXgEuB/TCPrWIB2g53Au9M8vYkm9JZbH1ln2PSKCQJnfVm91bVF7t2XQmc2Lw/Ebii17Fp9Krq01W1Q1VNo/P7d0NVfRTz2CpV9TjwcJJ3N02HAj/GPLbNL4APJJnY/I09lM76evPYIj4JqSWSfJDOGrQB4JtV9YU+h6RRSHIAcAuwiNfWDn6GzjrQ/wR2pPPH9NiqerovQeoNSXIwcGpVHZVka8xjqySZTudGsk2BB4CT6QzGmMcWSfJ54Dg63zSyAPhzYBLmsTUsQCVJktRTTsFLkiSppyxAJUmS1FMWoJIkSeopC1BJkiT1lAWoJEmSesoCVJJaJsmKJAuTLE5ycZKJq+j3w/V0vWOSnDGsbfaw7U2TzGuezS1Jq2UBKknt81JVTa+q3YDlwCndO5MMAFTVfuvpeqcB5zbn3j7J94C/aorgTzbXWg5cT+e7GSVptSxAJandbgF2TnJwkhuT/AedBx+Q5PmVnZKclmRRkruTnN20vSPJ1UnmJ7klyS7DT57kXcDLVfVU0zQLmE+nIN0buLqr++XACRvih5Q0tjhVIkkt1Ux3H8FrReA+wG5V9eCwfkcAxwDvr6oXk0xpds0BTqmqnyZ5P52i8pBhl9kfuKtrezmwNfB08xzue7v2LaZTlErSajkCKknts1mShcAQnUcOfqNpv2N48dk4DDi/ql4EqKqnk0wC9gMubs71VWC7EY7dDljatX0Onf8df5nk+ubRpDTnXQEsT7L5Ov10ksY8R0AlqX1eqqrp3Q1JAF5YRf8Aw5+7/FvAr4efZ6RrAZNXblTVM3SKz8eA7wNXJNmxqpY1Xd4MLPv/p5Gk1zgCKklj3zXAn668Wz7JlKp6FngwybFNW5LsMcKx9wI7r9xI8p4kK/93LAJeBd7U7NsaWNpMzUvSKlmAStIYV1VXA1cCQ810+6nNrhOAP0tyN3APcPQIh88D9kwzxEpnTegPgZOB24EvVNVzzb4ZwHc3zE8haSxJ1fBZGUmSXpPky8BVVXVdV9vsqpo9rN+lwKerakmPQ5TUMo6ASpLW5Cxg+Jfd39S9kWRT4HKLT0mj4QioJEmSesoRUEmSJPWUBagkSZJ6ygJUkiRJPWUBKkmSpJ6yAJUkSVJP/R8rvqN60KizVwAAAABJRU5ErkJggg==\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "# The next bit simply reorders the index by increasing average of weekday and weekend prices\n", - "# Compare the index order you get from\n", - "# state_price_means.index\n", - "# with\n", - "# state_price_means.mean(axis=1).sort_values(ascending=False).index\n", - "# See how this expression simply sits within the reindex()\n", - "(state_price_means.reindex(index=state_price_means.mean(axis=1)\n", - " .sort_values(ascending=False)\n", - " .index)\n", - " .plot(kind='barh', figsize=(10, 10), title='Average ticket price by State'))\n", - "plt.xlabel('Price ($)');" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "The figure above represents a dataframe with two columns, one for the average prices of each kind of ticket. This tells you how the average ticket price varies from state to state. But can you get more insight into the difference in the distributions between states?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.3.5.2 Distribution of weekday and weekend price by state" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Next, you can transform the data into a single column for price with a new categorical column that represents the ticket type." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 15#\n", - "#Use the pd.melt function, pass in the ski_data columns 'state', 'AdultWeekday', and 'Adultweekend' only,\n", - "#specify 'state' for `id_vars`\n", - "#gather the ticket prices from the 'Adultweekday' and 'AdultWeekend' columns using the `value_vars` argument,\n", - "#call the resultant price column 'Price' via the `value_name` argument,\n", - "#name the weekday/weekend indicator column 'Ticket' via the `var_name` argument\n", - "ticket_prices = pd.melt(ski_data[[___, ___, ___]], \n", - " id_vars=___, \n", - " var_name=___, \n", - " value_vars=[___, ___], \n", - " value_name=___)" - ] - }, - { - "cell_type": "code", - "execution_count": 20, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
stateTicketPrice
0AlaskaAdultWeekday65.0
1AlaskaAdultWeekday47.0
2AlaskaAdultWeekday30.0
3ArizonaAdultWeekday89.0
4ArizonaAdultWeekday74.0
\n", - "
" + "source": [ + "ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Y8Wtx5LzTNrz" + }, + "source": [ + "Weekend prices have the least missing values of the two, so drop the weekday prices and then keep just the rows that have weekend price." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "S2bTeAwbTNrz" + }, + "outputs": [], + "source": [ + "ski_data.drop(columns='AdultWeekday', inplace=True)\n", + "ski_data.dropna(subset=['AdultWeekend'], inplace=True)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "CuqQp0VITNrz", + "outputId": "58b61472-bebc-4f3a-b0f4-e6cf1ffec060" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(277, 25)" + ] + }, + "execution_count": 60, + "metadata": {}, + "output_type": "execute_result" + } ], - "text/plain": [ - " state Ticket Price\n", - "0 Alaska AdultWeekday 65.0\n", - "1 Alaska AdultWeekday 47.0\n", - "2 Alaska AdultWeekday 30.0\n", - "3 Arizona AdultWeekday 89.0\n", - "4 Arizona AdultWeekday 74.0" - ] - }, - "execution_count": 20, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ticket_prices.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This is now in a format we can pass to [seaborn](https://seaborn.pydata.org/)'s [boxplot](https://seaborn.pydata.org/generated/seaborn.boxplot.html) function to create boxplots of the ticket price distributions for each ticket type for each state." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 16#\n", - "#Create a seaborn boxplot of the ticket price dataframe we created above,\n", - "#with 'state' on the x-axis, 'Price' as the y-value, and a hue that indicates 'Ticket'\n", - "#This will use boxplot's x, y, hue, and data arguments.\n", - "plt.subplots(figsize=(12, 8))\n", - "sns.boxplot(x=___, y=___, hue=___, data=ticket_prices)\n", - "plt.xticks(rotation='vertical')\n", - "plt.ylabel('Price ($)')\n", - "plt.xlabel('State');" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Aside from some relatively expensive ticket prices in California, Colorado, and Utah, most prices appear to lie in a broad band from around 25 to over 100 dollars. Some States show more variability than others. Montana and South Dakota, for example, both show fairly small variability as well as matching weekend and weekday ticket prices. Nevada and Utah, on the other hand, show the most range in prices. Some States, notably North Carolina and Virginia, have weekend prices far higher than weekday prices. You could be inspired from this exploration to consider a few potential groupings of resorts, those with low spread, those with lower averages, and those that charge a premium for weekend tickets. However, you're told that you are taking all resorts to be part of the same market share, you could argue against further segment the resorts. Nevertheless, ways to consider using the State information in your modelling include:\n", - "\n", - "* disregard State completely\n", - "* retain all State information\n", - "* retain State in the form of Montana vs not Montana, as our target resort is in Montana\n", - "\n", - "You've also noted another effect above: some States show a marked difference between weekday and weekend ticket prices. It may make sense to allow a model to take into account not just State but also weekend vs weekday." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Thus we currently have two main questions you want to resolve:\n", - "\n", - "* What do you do about the two types of ticket price?\n", - "* What do you do about the state information?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.6.4 Numeric Features" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "Having decided to reserve judgement on how exactly you utilize the State, turn your attention to cleaning the numeric features." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.4.1 Numeric data summary" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 17#\n", - "#Call ski_data's `describe` method for a statistical summary of the numerical columns\n", - "#Hint: there are fewer summary stat columns than features, so displaying the transpose\n", - "#will be useful again\n", - "ski_data.___.___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Recall you're missing the ticket prices for some 16% of resorts. This is a fundamental problem that means you simply lack the required data for those resorts and will have to drop those records. But you may have a weekend price and not a weekday price, or vice versa. You want to keep any price you have." - ] - }, - { - "cell_type": "code", - "execution_count": 23, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0 82.424242\n", - "2 14.242424\n", - "1 3.333333\n", - "dtype: float64" - ] - }, - "execution_count": 23, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "missing_price = ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum(axis=1)\n", - "missing_price.value_counts()/len(missing_price) * 100" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Just over 82% of resorts have no missing ticket price, 3% are missing one value, and 14% are missing both. You will definitely want to drop the records for which you have no price information, however you will not do so just yet. There may still be useful information about the distributions of other features in that 14% of the data." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "#### 2.6.4.2 Distributions Of Feature Values" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Note that, although we are still in the 'data wrangling and cleaning' phase rather than exploratory data analysis, looking at distributions of features is immensely useful in getting a feel for whether the values look sensible and whether there are any obvious outliers to investigate. Some exploratory data analysis belongs here, and data wrangling will inevitably occur later on. It's more a matter of emphasis. Here, we're interesting in focusing on whether distributions look plausible or wrong. Later on, we're more interested in relationships and patterns." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 18#\n", - "#Call ski_data's `hist` method to plot histograms of each of the numeric features\n", - "#Try passing it an argument figsize=(15,10)\n", - "#Try calling plt.subplots_adjust() with an argument hspace=0.5 to adjust the spacing\n", - "#It's important you create legible and easy-to-read plots\n", - "ski_data.___(___)\n", - "#plt.subplots_adjust(hspace=___);\n", - "#Hint: notice how the terminating ';' \"swallows\" some messy output and leads to a tidier notebook" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What features do we have possible cause for concern about and why?\n", - "\n", - "* SkiableTerrain_ac because values are clustered down the low end,\n", - "* Snow Making_ac for the same reason,\n", - "* fastEight because all but one value is 0 so it has very little variance, and half the values are missing,\n", - "* fastSixes raises an amber flag; it has more variability, but still mostly 0,\n", - "* trams also may get an amber flag for the same reason,\n", - "* yearsOpen because most values are low but it has a maximum of 2019, which strongly suggests someone recorded calendar year rather than number of years." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.4.2.1 SkiableTerrain_ac" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 19#\n", - "#Filter the 'SkiableTerrain_ac' column to print the values greater than 10000\n", - "ski_data.___[ski_data.___ > ___]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Q: 2** One resort has an incredibly large skiable terrain area! Which is it?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 20#\n", - "#Now you know there's only one, print the whole row to investigate all values, including seeing the resort name\n", - "#Hint: don't forget the transpose will be helpful here\n", - "ski_data[ski_data.___ > ___].___" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**A: 2** Your answer here" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "But what can you do when you have one record that seems highly suspicious?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can see if your data are correct. Search for \"silverton mountain skiable area\". If you do this, you get some [useful information](https://www.google.com/search?q=silverton+mountain+skiable+area)." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "![Silverton Mountain information](images/silverton_mountain_info.png)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can spot check data. You see your top and base elevation values agree, but the skiable area is very different. Your suspect value is 26819, but the value you've just looked up is 1819. The last three digits agree. This sort of error could have occured in transmission or some editing or transcription stage. You could plausibly replace the suspect value with the one you've just obtained. Another cautionary note to make here is that although you're doing this in order to progress with your analysis, this is most definitely an issue that should have been raised and fed back to the client or data originator as a query. You should view this \"data correction\" step as a means to continue (documenting it carefully as you do in this notebook) rather than an ultimate decision as to what is correct." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 21#\n", - "#Use the .loc accessor to print the 'SkiableTerrain_ac' value only for this resort\n", - "ski_data.___[39, 'SkiableTerrain_ac']" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 22#\n", - "#Use the .loc accessor again to modify this value with the correct value of 1819\n", - "ski_data.___[39, 'SkiableTerrain_ac'] = ___" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 23#\n", - "#Use the .loc accessor a final time to verify that the value has been modified\n", - "ski_data.___[39, 'SkiableTerrain_ac']" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**NB whilst you may become suspicious about your data quality, and you know you have missing values, you will not here dive down the rabbit hole of checking all values or web scraping to replace missing values.**" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What does the distribution of skiable area look like now?" - ] - }, - { - "cell_type": "code", - "execution_count": 30, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "ski_data.SkiableTerrain_ac.hist(bins=30)\n", - "plt.xlabel('SkiableTerrain_ac')\n", - "plt.ylabel('Count')\n", - "plt.title('Distribution of skiable area (acres) after replacing erroneous value');" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You now see a rather long tailed distribution. You may wonder about the now most extreme value that is above 8000, but similarly you may also wonder about the value around 7000. If you wanted to spend more time manually checking values you could, but leave this for now. The above distribution is plausible." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.4.2.2 Snow Making_ac" - ] - }, - { - "cell_type": "code", - "execution_count": 31, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "11 3379.0\n", - "18 1500.0\n", - "Name: Snow Making_ac, dtype: float64" - ] - }, - "execution_count": 31, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data['Snow Making_ac'][ski_data['Snow Making_ac'] > 1000]" - ] - }, - { - "cell_type": "code", - "execution_count": 32, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
11
NameHeavenly Mountain Resort
RegionSierra Nevada
stateCalifornia
summit_elev10067
vertical_drop3500
base_elev7170
trams2
fastEight0
fastSixes2
fastQuads7
quad1
triple5
double3
surface8
total_chairs28
Runs97
TerrainParks3
LongestRun_mi5.5
SkiableTerrain_ac4800
Snow Making_ac3379
daysOpenLastYear155
yearsOpen64
averageSnowfall360
AdultWeekdayNaN
AdultWeekendNaN
projectedDaysOpen157
NightSkiing_acNaN
\n", - "
" + "source": [ + "ski_data.shape" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "uClN6iMtTNrz" + }, + "source": [ + "Perform a final quick check on the data." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zRCGGKFpTNrz" + }, + "source": [ + "### 2.11.1 Number Of Missing Values By Row - Resort" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "gVRW1q23TNr0" + }, + "source": [ + "Having dropped rows missing the desired target ticket price, what degree of missingness do you have for the remaining rows?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "23sQdjQlTNr0", + "outputId": "2fd71f30-f2ab-477e-be75-32db0b3db021" + }, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
count%
329520.0
62520.0
141520.0
86520.0
74520.0
146520.0
184416.0
108416.0
198416.0
39416.0
\n", + "
" + ], + "text/plain": [ + " count %\n", + "329 5 20.0\n", + "62 5 20.0\n", + "141 5 20.0\n", + "86 5 20.0\n", + "74 5 20.0\n", + "146 5 20.0\n", + "184 4 16.0\n", + "108 4 16.0\n", + "198 4 16.0\n", + "39 4 16.0" + ] + }, + "execution_count": 61, + "metadata": {}, + "output_type": "execute_result" + } ], - "text/plain": [ - " 11\n", - "Name Heavenly Mountain Resort\n", - "Region Sierra Nevada\n", - "state California\n", - "summit_elev 10067\n", - "vertical_drop 3500\n", - "base_elev 7170\n", - "trams 2\n", - "fastEight 0\n", - "fastSixes 2\n", - "fastQuads 7\n", - "quad 1\n", - "triple 5\n", - "double 3\n", - "surface 8\n", - "total_chairs 28\n", - "Runs 97\n", - "TerrainParks 3\n", - "LongestRun_mi 5.5\n", - "SkiableTerrain_ac 4800\n", - "Snow Making_ac 3379\n", - "daysOpenLastYear 155\n", - "yearsOpen 64\n", - "averageSnowfall 360\n", - "AdultWeekday NaN\n", - "AdultWeekend NaN\n", - "projectedDaysOpen 157\n", - "NightSkiing_ac NaN" - ] - }, - "execution_count": 32, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data[ski_data['Snow Making_ac'] > 3000].T" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You can adopt a similar approach as for the suspect skiable area value and do some spot checking. To save time, here is a link to the website for [Heavenly Mountain Resort](https://www.skiheavenly.com/the-mountain/about-the-mountain/mountain-info.aspx). From this you can glean that you have values for skiable terrain that agree. Furthermore, you can read that snowmaking covers 60% of the trails." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What, then, is your rough guess for the area covered by snowmaking?" - ] - }, - { - "cell_type": "code", - "execution_count": 33, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "2880.0" - ] - }, - "execution_count": 33, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - ".6 * 4800" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This is less than the value of 3379 in your data so you may have a judgement call to make. However, notice something else. You have no ticket pricing information at all for this resort. Any further effort spent worrying about values for this resort will be wasted. You'll simply be dropping the entire row!" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.4.2.3 fastEight" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Look at the different fastEight values more closely:" - ] - }, - { - "cell_type": "code", - "execution_count": 34, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.0 163\n", - "1.0 1\n", - "Name: fastEight, dtype: int64" - ] - }, - "execution_count": 34, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data.fastEight.value_counts()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Drop the fastEight column in its entirety; half the values are missing and all but the others are the value zero. There is essentially no information in this column." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 24#\n", - "#Drop the 'fastEight' column from ski_data. Use inplace=True\n", - "ski_data.drop(columns=___, inplace=___)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What about yearsOpen? How many resorts have purportedly been open for more than 100 years?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 25#\n", - "#Filter the 'yearsOpen' column for values greater than 100\n", - "ski_data.___[ski_data.___ > ___]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Okay, one seems to have been open for 104 years. But beyond that, one is down as having been open for 2019 years. This is wrong! What shall you do about this?" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "What does the distribution of yearsOpen look like if you exclude just the obviously wrong one?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 26#\n", - "#Call the hist method on 'yearsOpen' after filtering for values under 1000\n", - "#Pass the argument bins=30 to hist(), but feel free to explore other values\n", - "ski_data.___[ski_data.___ < ___].hist(___)\n", - "plt.xlabel('Years open')\n", - "plt.ylabel('Count')\n", - "plt.title('Distribution of years open excluding 2019');" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The above distribution of years seems entirely plausible, including the 104 year value. You can certainly state that no resort will have been open for 2019 years! It likely means the resort opened in 2019. It could also mean the resort is due to open in 2019. You don't know when these data were gathered!" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Let's review the summary statistics for the years under 1000." - ] - }, - { - "cell_type": "code", - "execution_count": 38, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "count 328.000000\n", - "mean 57.695122\n", - "std 16.841182\n", - "min 6.000000\n", - "25% 50.000000\n", - "50% 58.000000\n", - "75% 68.250000\n", - "max 104.000000\n", - "Name: yearsOpen, dtype: float64" - ] - }, - "execution_count": 38, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data.yearsOpen[ski_data.yearsOpen < 1000].describe()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The smallest number of years open otherwise is 6. You can't be sure whether this resort in question has been open zero years or one year and even whether the numbers are projections or actual. In any case, you would be adding a new youngest resort so it feels best to simply drop this row." - ] - }, - { - "cell_type": "code", - "execution_count": 39, - "metadata": {}, - "outputs": [], - "source": [ - "ski_data = ski_data[ski_data.yearsOpen < 1000]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "##### 2.6.4.2.4 fastSixes and Trams" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "The other features you had mild concern over, you will not investigate further. Perhaps take some care when using these features." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.7 Derive State-wide Summary Statistics For Our Market Segment" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You have, by this point removed one row, but it was for a resort that may not have opened yet, or perhaps in its first season. Using your business knowledge, you know that state-wide supply and demand of certain skiing resources may well factor into pricing strategies. Does a resort dominate the available night skiing in a state? Or does it account for a large proportion of the total skiable terrain or days open?\n", - "\n", - "If you want to add any features to your data that captures the state-wide market size, you should do this now, before dropping any more rows. In the next section, you'll drop rows with missing price information. Although you don't know what those resorts charge for their tickets, you do know the resorts exists and have been open for at least six years. Thus, you'll now calculate some state-wide summary statistics for later use." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Many features in your data pertain to chairlifts, that is for getting people around each resort. These aren't relevant, nor are the features relating to altitudes. Features that you may be interested in are:\n", - "\n", - "* TerrainParks\n", - "* SkiableTerrain_ac\n", - "* daysOpenLastYear\n", - "* NightSkiing_ac\n", - "\n", - "When you think about it, these are features it makes sense to sum: the total number of terrain parks, the total skiable area, the total number of days open, and the total area available for night skiing. You might consider the total number of ski runs, but understand that the skiable area is more informative than just a number of runs." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "A fairly new groupby behaviour is [named aggregation](https://pandas-docs.github.io/pandas-docs-travis/whatsnew/v0.25.0.html). This allows us to clearly perform the aggregations you want whilst also creating informative output column names." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 27#\n", - "#Add named aggregations for the sum of 'daysOpenLastYear', 'TerrainParks', and 'NightSkiing_ac'\n", - "#call them 'state_total_days_open', 'state_total_terrain_parks', and 'state_total_nightskiing_ac',\n", - "#respectively\n", - "#Finally, add a call to the reset_index() method (we recommend you experiment with and without this to see\n", - "#what it does)\n", - "state_summary = ski_data.groupby('state').agg(\n", - " resorts_per_state=pd.NamedAgg(column='Name', aggfunc='size'), #could pick any column here\n", - " state_total_skiable_area_ac=pd.NamedAgg(column='SkiableTerrain_ac', aggfunc='sum'),\n", - " state_total_days_open=pd.NamedAgg(column=__, aggfunc='sum'),\n", - " ___=pd.NamedAgg(column=___, aggfunc=___),\n", - " ___=pd.NamedAgg(column=___, aggfunc=___)\n", - ").___\n", - "state_summary.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.8 Drop Rows With No Price Data" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "You know there are two columns that refer to price: 'AdultWeekend' and 'AdultWeekday'. You can calculate the number of price values missing per row. This will obviously have to be either 0, 1, or 2, where 0 denotes no price values are missing and 2 denotes that both are missing." - ] - }, - { - "cell_type": "code", - "execution_count": 41, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0 82.317073\n", - "2 14.329268\n", - "1 3.353659\n", - "dtype: float64" - ] - }, - "execution_count": 41, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "missing_price = ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum(axis=1)\n", - "missing_price.value_counts()/len(missing_price) * 100" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "About 14% of the rows have no price data. As the price is your target, these rows are of no use. Time to lose them." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 28#\n", - "#Use `missing_price` to remove rows from ski_data where both price values are missing\n", - "ski_data = ski_data[___ != 2]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.9 Review distributions" - ] - }, - { - "cell_type": "code", - "execution_count": 43, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAA24AAAJOCAYAAADPilTlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdebwcVZn/8c+XhCWEsEYjSySKiIIosinjFsUFxBEcRUFUMi7oKKOMUQH1p+iIRkccXHBlVXYXIIMLInJBRPYt7AQIJCQQwpoEVALP749zOqnb6b63+95eqm9/369Xv253VXXVU3WfPn3OqVPVigjMzMzMzMysvNbodgBmZmZmZmY2NDfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NtzoknSjpa61etp2ajHmGpEvaHZN13ljP3VFuZ0DSh9u9HbN2kvR7SQd2O45+J2mZpOc3uGxIekGLtjstr298jXnPzXGNy6+dK2ZjSF823HLl7RFJa7dh3dMlLSi83jQXsFMK075QZ9ofWh2PjS3OXesVkuZJemO34xhKdc7naUdIeipXfh+VdKmk3boVYy0RsWdEnNTtOMa6nMMPSJpYmPZhSQMAEbFeRNzVgu2s1pEqaQtJv5a0RNJjkuZImjHcuiLi3hzX0/m1c8U6Jn9mnszl5/25U3a9bsc1lvRdw03SNOA1QABvb/f2ImIRMBd4bWHya4Fba0y7uN3xWO9y7pp1zBkRsR4wGbgQ+GWX47HuGQ98qgvb/QUwH9gS2AT4APBAF+Iwa9a/5vJzB+DlwOFdjmdM6buGG6nwuww4EVg5fEDSyyVdI2mppDOAdQrzavWGrTbsIffK/R7YLPc2LJO0GalS+9q8zDhSIn+3atpueTkkfVDSLfnMynmStixs40WSzpf0sKTbJL271k5KmiTpQknfU7KJpNmSHpd0BbBV1fLflTQ/z79a0mvy9OdIekLSJoVld5L0oKQ1Gzje1jr9mrt135d7846R9Nu8/5dL2qow/02Sbs091j8A1PDRtpaTtLakoyUtzI+jlc8eV85+SZopabGkRZL+vfDeTST9Xy6jrpT0tWJuD5Mnb5V0c86R+yR9ZoicXykiVgCnAJtLelZe16AziUpn6E7OzytD2A6UdK/S2ZIvNHBcjpD0S0kn5xjnSHqhpMPzsZgv6c2F5T3kt3P+B/iMpA2rZxTL0uHyM3ujpDty+XhMLt9eDPwY2C3n4KN52V2AEyNieUSsiIhrI+L3tQKU9M6cly9R1TDKYq5Uvg8kfTvHcLekPQvreZ6ki3MO/inHePJwByjn7v25nL1Y0naFeRMkHSXpnjz/EkkThlun9b6IuB84D9hBtUc3rCxLcxl4pqSf5/y7SdLOhWUPzWX30ly+797ZvSmPfm24nZIfb5E0RdJawNmkHq6NSb2r72x2xRGxHNgTWJiHKqwXEQspVH5JFd9bgQuqpq0JXCFpH+DzwL8BzwL+ApwGKyvX5wOnAs8G9gd+WCwk83Kb5PX/NSI+GREBHAP8HdgU+GB+FF1J6h3ZOK//l5LWyR+8AaBYyX4fcHpEPNXsMbJR6bvcBdZt4H37A18BNiKdITwyr2sy8Gvgi6QzJ3cCr2r22FhLfQF4JamseRmwK+n/U/EcYANgc+BDwDGSNsrzjgGW52UOZHDnxXD5dRzw0YiYBLwE+PMQOU9hvWuRPncPAY80sZ+vBrYBdge+lCvnw/lX0ud4I+BaUoVnjXwsvgr8pIntW+tcRfoO/Mwwy9XNz4K3kRpkLyN9p74lIm4BPgb8LedgpYF4GSn/95P03HobVerc+Cbwxoi4sYH9eQVwG6lM/BZwnKRKh9apwBWkM3xHAO9vYH2QOkC2Jn32riF9R1V8G9gJ+BfSd9TngGcaXK/1MElbkMrYuQ2+5e3A6cCGwGzgB3k92wAHA7vkMvwtwLxWx9sr+qrhJunVpGEHZ0bE1aSK3HtJFYk1gaMj4qmI+BWpIdMqFwEvyRWQ1wB/iYg7gMmFaZdFxD+BjwLfiIhbcm/v10m9FVuSCv15EXFC7oG7hlQxfVdhW5vl7f0yIr6Y93scqTL/pdx7dyMwaMx7RJwcEQ/l9R4FrE2qeJCXfV9hXfuTKhjWIf2auw2+7zcRcUXh7MgOefpbgZsj4le5k+Fo4P4WHhtr3gHAVyNicUQ8SGpwFyuHT+X5T0XE74BlwDaFMuzLEfFERNzM4DJsuDx5CthW0voR8UieP5R35zMfTwIfAd6V86tRX4mIJyPieuB6UkV9OH+JiPPydn5J6vyYlXP3dGBarbM+1hFfAv5T+axrtQbys2JWRDwaEfeShuDuUGOZin1JnV//D7hb0nWSdqla5hDgs8D0iGi0cnxPRPwsXwN3Eqkzd0puHO5Cqif8MyIuIVWehxURx0fE0oj4B6nB9zJJG0hag9RJ/KmIuC8ino6IS/NyNnadLWkpaajvYuDLDb7vkoj4Xc7NX7Cq3HyaVCfdVtKaETEvIu5sedQ9oq8abqQesD9GxJL8+tQ8bTPgvnxmquKeVm00IuYBC0i9sK8lFcYAfytMq1wjtCXwXaWL4h8FHiYN79o8z3tFZV6efwCph69iL2ACaehFxbNI4/Tn19s/peFJt+ShDI+Ser0n59nnkD4wzwfeBDwWEVeM6GDYSPVr7jbyvmJj7AmgciH0ZhRyPh+j4mfAOm8zBufnPXlaxUNVDaTK/7NWGVZ8PlyevJPUkL9H0kUa/mYjZ+YzH1OAG0lnDJpRLyeHUrx+6UlgSeUGE/k1Da7HWix3dp4LHFZnkeHys6LhvMgdDIdFxHakPLyOVCEuDvf+LHBMRCyouZLaVsYQEU/kp+uRPocPF6bV24dBJI2TNEvSnZIeZ9WZkMn5sQ6po9H6xz75zNh04EWsqksOp/rzsY6k8blT4hBSp8BiSaeralh7P+mbhlseU/1u4HV5LPb9wH+RWvSLSNcwFAvE4tCE5aQhW5V1FSuN1aLO9L+QKrm7AZdWTXs1qyq/80lDejYsPCZExKV53kVV89aLiP8obOdnwB+A32nVnbAeBFYAU2vtn9L1bIeSjs9GucLyGPl6oIj4O3AmqSL0fny2raP6PHcbeV89iyjkfD5GU+svbh2wkNTIqnhunjacShm2RWFa8X85ZJ5ExJURsTdpKNfZpPIM6uc8+X1LSGeSj5C0aZ486DPF4E4EG7u+TDr7unmNecPl53AaycNvkxpXGxdmvRn4oqSmh8fXsAjYWFIxtxvZh/cCewNvJHX4TsvTBSwhXaKxVc132pgWEReRrsn/NqvXRcaROjwaXdepEVEZeRSk4cF9qW8absA+pNOt25KGJ+wAvJhUAd2HVOh+UtJ4Sf9Guvai4npgO0k7SFqH1Oqv5wFgE0kbVE2/mHStxMKIeDxPuyRP24B0BgPS2YbDK9dm5OEG++Z55wIvlPR+SWvmxy41rp84mDSG/VxJE3Kv7W9IlY91JW3L4PH3k/L+PwiMl/QlYP2qdf4cmEEagzzsxcrWUn2bu028r5bf5n3/N6UL9T+JK9mdtqakdSoP0jWPX5T0LKVrEL9EA+VJjTLsRaT8q6ibJ5LWknSApA3ysMPHSZ8nqJ/zxW3fSrre7HN50nXAfnkbOzN42K6NUbnX/wxSOVI9b7j8HM4DwBZK11QCIOmbSjcbGS9pEvAfwNyIeKjwvpuAPUjXwo3qTsMRcQ/per4j8mdmN9J1l8OZBPyDdB3ouqQh8pV1PgMcD3xH0mb57NxuasPP2VhpHU0aqbUu6QzaXko3tvsiafjjsCRtI+kNOW/+ThqB8PQwbxuz+qnhdiBwQqTfOLm/8iBd/Lg/6YYKM0gXoL+HVAgDEBG3ky4O/xNwB6nSWlP+kj8NuCsP2amczr2I1NtbfO91pKFhV1eGJ0TEWaSehNPzsIMbSRd3EhFLST1s+5F6qe/Pyw5K/jwk7CBSL/Q5ucJ0MGk4xP2kHpATCm85j3Rx8e2koUt/p2qIRET8lXRB8TV5+Jx1Tt/mLunapGHfV2d/lpCuE5lFqlRsDfx1uPdZS/2O9CVbeaxDqhzeAMwh3cig0R9eP5jUUXA/6az/aaQKYyP59X5gXs7Lj5Gv2R0i56v9D3CQpGeTrjnaivR5+wpp2LL1h68CE+vMq5ufDfgzqRF2v6TKcPh1gbOAR4G7SGcaVmuc5eso3wb8TIU7RI7QAaSRFQ+RPpdnMPw+/JxUb7gPuJl0U5Wiz5A+61eShs9/k/6qe/a1SNcy/5zU4fFx4FhSriwnXYbRiLVJ3+NLSJ+vZ5NuhNaXNPjSGLP6JP0ZODUiju12LGbW3yR9E3hORNS6e59ZV42F/FT6eZlbI6LRm0uYWZu518MaonQ3qx1JPXBmZh2l9DttL1WyK+nnAs7qdlxmMDbyMw8v3krSGpL2IF27dna34zKzVdxws2FJOok01O6QPCTJzKzTJpGGAS8n3VzkKNJw2p4g6fda9UPfxUffDvkZY3o6P7PnkH6zbhnwPeA/IuLafI1ordy9qavRmvUhD5U0MzMzMzMrOZ9xMzMzMzMzK7nx3Q4AYPLkyTFt2rSuxrB8+XImTqx3s6jucExw9dVXL4mIhn/roxu6mb9lzBFwXBVlz996uVvW/1+RY2yNejGWPXeht/K3jDHB2I2r7PlbK3fL+L9wTI1pZUzD5m5EdP2x0047RbddeOGF3Q5hNY4pArgqSpCjQz26mb9lzJEIx1VR9vytl7tl/f8VOcbWqBdj2XM3eix/yxhTxNiNq+z5Wyt3y/i/cEyNaWVMw+Wuh0qamZmZmZmVnBtuZmZmZmZmJVeKa9xaadphv21q+Xmz9mpTJGbNaTZ3AU7co1zjvK1/uey1sphz32PMaCIfnYtWFiOpBzh/+4vPuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZyY25a9yaVRlPPHP7FQ2NifdYYjMzMzMz67S+b7iZmdlgzd7cwczMzNrPQyXNzMzMzMxKzg03MzMzMzOzknPDzczMzMzMrOR8jZtZD/MPzZqZmZn1B59xMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzEpufLcDMLPOmXbYb5taft6svdoUiZmZmZk1w2fczMzMzMzMSs4NNzMzMzMzs5Ib1VBJSfOApcDTwIqI2FnSxsAZwDRgHvDuiHhkdGGamVmFy14zM7P+04pr3F4fEUsKrw8DLoiIWZIOy68PbcF2zMxsFZe91pPK1vHQ7LW/4Ot/xzJJxwNvAxZHxEvytLr5Kelw4EOkfP5kRJzXhbCtT7RjqOTewEn5+UnAPm3YhpmZDeay13rJ6yNih4jYOb+udDxsDVyQX5t1w4nAHlXTauanpG2B/YDt8nt+KGlc50K1fjPaM24B/FFSAD+JiJ8CUyJiEUBELJL07FpvlHQQcBDAlClTGBgYGGUoycztV4zofVMmNPbeVsXZiGXLlnV0e40oY0wjVbZeX7MmtLXsbbQ8HI3RliO9UBY5xqbsDUzPz08CBvAZY+uCiLhY0rSqyfXyc2/g9Ij4B3C3pLnArsDfOhGr9Z/RNtxeFRELcwXhfEm3NvrGXNH4KcDOO+8c06dPH2UoyYwRDHmAVEk5ak4Dh2PO8qbXPdIhFQMDA7TquLRKGWMaJQ83s17U1rL3+6ec01h5OArzDlh9u83ohbLIMdbVdx0PJWogD+K4GlYvPzcHLisstyBPW81wubts2TJmbv9004G18ziV8P/Q9zGN6ps5Ihbmv4slnUXqZXhA0qY5sTcFFrcgTrNOca+vld5YKHv9m4J9re86HsraiHdco6Ya06LWgsPl7sDAAEddMoKTA6PsBBtKGf8P/R7TiEs2SROBNSJiaX7+ZuCrwGzgQGBW/ntOKwI1a4NSDfUdSQ9xu3uWR7pfZewRg/LG1QyXvdbrxkLHg/Wdevm5AJhaWG4LYGHHo7O+MZouqSnAWZIq6zk1Iv4g6UrgTEkfAu4F9h19mGZtUaqhviMZ5tvwEN8RGmlPXhl7xKC8cTXJZa/1LHc8WI+ql5+zgVMlfQfYDNgauKIrEVpfGHGNLyLuAl5WY/pDwO6jCcqsE9zra73IZa/1OHc8WKlJOo10ycRkSQuAL5MabKvlZ0TcJOlM4GZgBfCJiGj+QjWzBrV3ELhZSbnX18ys89zxYGUXEfvXmVUzPyPiSODI9kVktoobbtav3OtrZmZmZj3DDTfrS+71NTMzM7Neska3AzAzMzMzM7OhueFmZmZmZmZWch4qaWZmZmbWg6Y1+VNC82bt1aZIrBN8xs3MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzkfFdJMzMrveo7p83cfgUzhrmbmu+eZmZmY4nPuJmZmZmZmZVcqc+4NfvbFGZmZmZmZmORz7iZmZmZmZmVXKnPuJlZdzV71tvXFJmZmZm1h8+4mZmZmZmZlZwbbmZmZmZmZiXnhpuZmZmZmVnJ+Rq3DvB1QmZmZmZmNho+42ZmZmZmZlZybriZmZmZmZmVnBtuZmZmZmZmJeeGm5mZmZmZWcn55iRmZmZmTWj2pmMn7jGxTZGYWT9xw83MzMakZivXzfIdgM3MrJM8VNLMzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5Jzw83MzMzMzKzkfFdJszZp9x3tzMzMzKx/uOFmZmZmZtYHmulUnrn9Cqa3LxQbAQ+VNDMzMzMzKzmfcTMzMzNrozn3PcaMJs50+MfdzawWN9xKqHIae+b2Kxoq6F3AW1k0m7vg/DUzMzNrhBtuY0CzN8FwRdnKxPlrZmZWTiO50Zq/p9vH17iZmZmZmZmVnBtuZmZmZmZmJde2oZKS9gC+C4wDjo2IWe3allkrOXetlzl/rVc5d62XOX9X8SUQ7dOWhpukccAxwJuABcCVkmZHxM3t2J41ZzQ/DN3MTSeaUZYPrXO3/Jy/9Tl/O2u4XGxXvrXSiXtM7HYIgHO32lio+HairC7Lfjt/rVPadcZtV2BuRNwFIOl0YG/ACWxl59y1Xub8tV7l3O2waYf91ncAbh3n7yiMhR8F71RniyJiRG8ccqXSu4A9IuLD+fX7gVdExMGFZQ4CDsovtwFua3kgzZkMLOlyDNUcE2wZEc/q1MYayd08vSz5W8YcAcdVUbr8bTB3y/r/K3KMrVEvxtLlbp7eq/lbxphg7MZVuvxtIHfL+L9wTI1pZUxD5m67zripxrRBLcSI+Cnw0zZtv2mSroqInbsdR5Fj6ophcxfKk79l/X84rq5pSdnbC8fJMbZGiWJsWdlbon1aqYwxgeNqoVGXvWXcZ8fUmE7G1K67Si4AphZebwEsbNO2zFrJuWu9zPlrvcq5a73M+Wsd0a6G25XA1pKeJ2ktYD9gdpu2ZdZKzl3rZc5f61XOXetlzl/riLYMlYyIFZIOBs4j3Rb1+Ii4qR3baqGuD3urwTF1WA/mbln/H46rC1qYv71wnBxja5QixhaXvaXYpypljAkcV0u0KH/LuM+OqTEdi6ktNycxMzMzMzOz1mnXUEkzMzMzMzNrETfczMzMzMzMSq4vG26S5kmaI+k6SVflaRtLOl/SHfnvRh2I43hJiyXdWJhWNw5Jh0uaK+k2SW/pYExHSLovH6/rJL21kzFZ4rwdUVzO3SZJ2iMfk7mSDut2PNUkTZV0oaRbJN0k6VPdjqkeSeMkXSvp3G7HUoukDSX9StKt+Xju1u2YRqObuVsvL0tSNg7Kw5LEtFrulSGubulW7pbt+7yMnyNJ60i6QtL1OaavdDWmiOi7BzAPmFw17VvAYfn5YcA3OxDHa4EdgRuHiwPYFrgeWBt4HnAnMK5DMR0BfKbGsh2JyY+Vx9t523xczt3mjuG4fCyeD6yVj9G23Y6rKsZNgR3z80nA7WWLsRDrp4FTgXO7HUud+E4CPpyfrwVs2O2YRrEvXc3denlZkrJxUB6WJKbVcq8McfVb7pbt+7yMnyPSb/Stl5+vCVwOvLJbMfXlGbc69iYVJOS/+7R7gxFxMfBwg3HsDZweEf+IiLuBucCuHYqpno7EZENy3g4dVz3O3dp2BeZGxF0R8U/gdNKxKo2IWBQR1+TnS4FbgM27G9XqJG0B7AUc2+1YapG0PqnSdhxARPwzIh7tblSj0tXcHSIvu1o21snDbsdUL/e6/j3SJV3L3bJ9n5fxcxTJsvxyzfyIbsXUrw23AP4o6WpJB+VpUyJiEaTEAZ7dpdjqxbE5ML+w3AI6W1k5WNIN+bR65XRwt2PqN87bkXHuNq6njoukacDLST2gZXM08DngmW4HUsfzgQeBE/IwumMlTex2UKNQmtytystul4218rDbMdXLvW7H1S1l279S/B/K9DnKw42vAxYD50dE12Lq14bbqyJiR2BP4BOSXtvtgBqgGtM69VsOPwK2AnYAFgFHlSCmfuS8bZ5ztzk9c1wkrQf8GjgkIh7vdjxFkt4GLI6Iq7sdyxDGk4ZI/SgiXg4sJw336VWlyN0m8rLt8Y4gDzt1DJvNvVL8b9uoV/avY3GW6XMEEBFPR8QOwBbArpJe0q2Y+rLhFhEL89/FwFmkU5gPSNoUIP9d3KXw6sWxAJhaWG4LYGEnAoqIB3LSPgP8jFWnfLsWUz9y3jbPudu0njguktYkfamfEhG/6XY8NbwKeLukeaRhT2+QdHJ3Q1rNAmBB7jkG+BWpMt2rup67dfKym2VjvTzsdnldL/e6HVe3lG3/uvp/KOHnaKU8pHcA2KNbMfVdw03SREmTKs+BNwM3ArOBA/NiBwLndCfCunHMBvaTtLak5wFbA1d0IqBKYmbvIB2vrsbUb5y3I+PcbdqVwNaSnidpLWA/0rEqDUkiXRtzS0R8p9vx1BIRh0fEFhExjXQM/xwR7+tyWINExP3AfEnb5Em7Azd3MaTR6mruDpGXXSsbh8jDrpbXQ+ReKb9HOqBs5W7X/g9l/BxJepakDfPzCcAbgVu7FlOr7nLSKw/S2Orr8+Mm4At5+ibABcAd+e/GHYjlNNLwradILfQPDRUH8AXS3WluA/bsYEy/AOYAN+SE3LSTMfnhvB1FXM7d5o/jW0l38rqzkmdlegCvJg07uQG4Lj/e2u24hoh3OuW9q+QOwFX5WJ4NbNTtmEa5P13L3Xp5WYayMW9rZR6WIaZauVeGuLr16Fbulu37vIyfI+ClwLU5phuBL+XpXYlJeQNmZmZmZmZWUn03VNLMzMzMzKzXuOFmZmZmZmZWcm64mZmZmZmZlZwbbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcm64mZmZmZmZlZwbbmbWdpJC0gvqzFsm6fmdjsn6l6SbJE3vwnYHJH2409s16wRJ75A0P5fpLx9m2RmSLim8rvsdYWarjPmGW3XhUDXvAEl/bHA9R0g6eYj58yS9caRxtpuk5+bCdFy3Y7H6JL1a0qWSHpP0sKS/Stqli/EM5C/Ul1VNPztPnz7abUTEehFx12jXY70ll0eVxzOSniy8PqCd246I7SJioME45xVie0DSCZLWa2d8Zs2StJakoyQtyLl6t6T/7XAY3wYOzmX6tR3etrVI2euzAJJOlPS1bsfRDWOm4TaSCm9EnBIRb+5gjK8pVEyW54pvsfLy3HZtOyLuzYXp0+3aho2OpPWBc4HvAxsDmwNfAf7RzbiA24EPVF5I2gR4JfBg1yKynpfLo/UiYj3gXuBfC9NOaWQdksY3Mq0F/jXHuSOwC/DFZt6sZMx831rz2pSXRYcDOwO7ApOA1wOdbjxtCdzU4W2a9ZUx8UVS4grvIBHxl0JFZbs8ecNCZeXeRtZTp7LiM2m974UAEXFaRDwdEU9GxB8j4gZYdfZY0rclPZJ7VPesvFnSZpJm546LuZI+kqevk88YTM6vvyhpRf7cIOlrko4eIq5TgPcUcmx/4Czgn4Vt7yrpb5IelbRI0g8krVVrZbmTZb6k1+fXK4fI5F60YyT9VtJSSZdL2qrw3jdLui130PxQ0kXDDT2TtJWkP0t6SNISSadI2rAwf6qk30h6MC/zg6HWZ+0laQ1Jh0m6M/8/zpS0cZ43LefLhyTdC/w5fy7+Kul/JT0MHNHA/3xlj7LSaIozJf0859xNknauFVtE3Af8HniJpI0knZvz5pH8fIvCNgYkHSnpr8ATwKDhwJI2lXSDpM/k1zMk3ZVjuFttPutogxVybqmkm5WG/a2dy7SXFJZ7Vi5Pn51fv03SdXm5SyW9tLDsPEmHSroBWC5pfK3tFJYfp3TWbEnOgYNzvo/P8zeQdFwuY+/LZXelXN4FOCsiFkYyLyJ+XhXLZ3LOPSbpDEnrFOZ/ROl742Gl75HN8vSvSPp+fr6mUqfzt/LrCZL+LmmKpGXAOOB6SXfWO6Yt/reZ9Z0x0XBjmApvkaT/Uar8bqDVx1h/V6lC+bikqyW9purt6+TCbqmka1Q1fKywnroVj3qGKpDrVExOlPQjSb+TtBx4vaS9JF2b458v6YjC+isVnsoXwICk/87rXSrpj8oV+2Hi/KWk+3PBf7Gk7QrzJuQvnXvy/EskTRhunbbS7cDTkk6StKekjWos8wrgNmAy8C3gOEnK804DFgCbAe8Cvi5p94j4O3Al8Lq83GuBe4BXFV5fNERcC4GbgcrZ6Q8AP69a5mngv3JcuwG7Ax+vXpGkt+Q43xkRF9bZ3v6kjpeNgLnAkfm9k4FfkXqWN8nH4V+GiHvlZoFvkI7Li4GpwBF5neNInT73ANNInT6nN7BOa59PAvuQ8nUz4BHgmKplXkf6X74lv34FcBfwbFK+1P2f1/F20v99Q2A2ULPxLmkq8FbSmYw1gBNIZxmeCzxZ433vBw4inQG5p7CeaaTP3A8i4tuSJgLfA/aMiEmkvL5uiHit9e4EXgNsQCp/TiZ1BP+GVCZVvBu4KCIWS9oROB74KKlM+gkwW9LaheX3B/YiddKuqLUdSZvmZT8C7AnsQDq7u09VjCcBK4AXAONkdEgAACAASURBVC8nlcmVjqvLgE9L+rik7QvfC0XvBvYAnge8FJgBIOkNpM/Lu4FNSblaKQcvAqbn57sA97Pqu2Q34LaIeCB3SAO8LCIqnW1D7auV2y65sf2I0vDwdTR8Z1XdzidJH5R0S37feZK2HC4ASS+SdH7uTLhN0ruHWLZmB0qui/+qatnvSvreyA5LCUREzz+A9YGHSIXansBGhXkzgEtIX7I/A84D1i3OKyz7PlLhOx6YSSqg1snzjgCeIlWI1wQ+A9wNrJnnzwPemJ8fQipEtwDWJhXmp1XFPA0IYHx+fXZebiKp8nEF8NFCnCuA/8yxTQBOBB4jVb7XANYhFa7b59cvBR4A9qmzvQFSofrCvL4BYFYDx/qDpErI2sDRwHWFecfk9WxO6nn7F2DtbudHLz1IlcwTSQ2wFaRK5JRCHswtLLtu/p8+h1QxfRqYVJj/DeDE/Py/SRXD8TmvPwXMynnzJDC5TjwDpIrB+0gNrm2A2/O8BcD0Ou87hNT7W3kdpAbXPcD2VcsG8IL8/ETg2MK8twK35ucfAP5WmCdgPvDhJo/xPsC1+flupCGf47v9v+/nB4PLz1uA3QvzNiWVveML5djzC/NnAPc2+j+vsb0jgD8V5m0LPFm17DLg0Zy/PwQm1NjGDsAjhdcDwFerlhkAvpPXuX9h+sS8/nfWWrcfXcnJ64C9gTcCdxWm/xX4QH7+I+C/q953G/C6Qu58sJHt5Od/Jn/v59dvzPk+HphCGkU0oTB/f+DC/Hwc8Ikc3z9IHW4HFpadB7yv8PpbwI/z8+OAbxXmrZc/c9NI9YO/k+pGhwGfJ5X965EaY98rvG9lWd7Avs5gcP1ryPf60dHcnwfcSKpXbJxz6ms5B95JqntMAn4JnJ3fMxF4HNgmv94U2C4/34fUCfvinMtfBC4dJoaJpO/3f8/v2RFYUljnicDX8vMdgcWkDrxxwIF5H9Ymdaw9Aawfqz4ni4BXdvs4j/QxJs64RcTjwKtJH/yfAQ8qneqfkhdZk1Tp3Jh0rcITddZzckQ8FBErIuIo0j99m8IiV0fEryLiKdKX7zqka32qfRT4QkQsiIh/kCoG71KdMe45zj2BQyJieUQsBv4X2K+w2MKI+H6O7ck87ZyI+GtEPBMRf4+IgYiYk1/fkPf5ddR3QkTcntd3JqniMaSIOD4ilhb262X5bOEapEbdpyLivkhnPi/Ny1mDIuKWiJgREVsALyGdMSgOY7y/sGwlj9fLyz0cEUsLy95DakTDql7THYE5wPmk3HglqTG4ZJjQfgO8gdR58IvqmZJemHvf7pf0OPB10tm3okOAMyNizjDbur/w/Im8f5D2cX5lRqRSeMEw60LSsyWdns9kP07qSa/ENhW4J1JPuJXDlsBZuef0UVJD7mlSxbViftV7Br0e5n9eS3XOrVNVXu8TERtGxJYR8fGIeFLSupJ+kkcYPA5cDGyowcPWq+MEOAC4j3T2GICIWA68B/gYsEhpqPCLhojXWkzSBwo99o+Syt/JpMbUBEmvyGcJdiANFYeUqzMr78nvm0oqqyqqc7PedqCqjKt6viWpLrOo8N6fkDp6yd+5x0TEq0hnjo8Ejpf04sI6hipbV54RjohlpM7wzXP94CrS90VldMalpE7j1zHEaI1h9tXK7QcRMT8iHibl0v65fvzriHgi1zWOZHAd8xnSMPIJEbEoIirXO34U+Eau36wg1Q92GOas29uAeRFxQq73XgP8mnTypNpHgJ9ExOX5c3ASqfPilRFxD3ANq85evwF4IiIuG8lBKYMx0XCDYSu8LyD1nH0lIv5Zbx2SZuZTuY/lQmYDBhcyxUrjM6wallatkYpH9fJ1C+Tqbdeblr9YLsynsR8jVQKaqawMeac0pfH3s5SGgD5O6tEgb2MyqSF751DrsMZFxK2kXqWXDLMopN7VjSVNKkx7LqmCCOmLdhvgHaRhPjfn+Xsx9DDJSixPkK7t+Q9qNNxIPc+3AltHxPqkXtnqoTr7AvtIOqSB/allEeksNpBu+FB8PYRvkDp1Xppje18htvnAc+t1qlhXzCcNGdyw8Fgn0vVlFVH1nurXQ/3PW2Um6TP1iryN1+bpxe1UxwWpw2sJcGqxkRcR50XEm0g91beSOiGtA3IF8mfAwcAmEbEh6YyD8nf9maSzW+8Fzi10kM0HjqzK1XUj4rTC6qOR7eRFBpVxpEZgxXxSZXRyYVvrR8R2VIl0ucgxpGHG2zZwCBaS6iGVOCeSzq5UPnMXkSq8LycNu7+INEx5V1KHxWoa2Fcrt2L98h5gs6E6q4bpfNoS+G6hfvswKQ82p74tgVdUdYocQBphVGvZoTpQTmXVcOf35tc9a8w03IpqVHhvIZ1u/b2kbWq9R+l6tkNJY7w3yoXMYwwuZKYWll+DVMAurLG6Rioe1csPVyDXqgBUTzuVNLRuakRsAPyY1haS72XV0JENSMMoyNtYQhpOsVXNd9qw8njumZUx40rX0+xPGnY7pIiYT2qcfUNpLPpLgQ+RbixSaXhdTRpKU2moXUrqCRu24ZZ9njQEaF6NeZNIwySW5cL6P2oss5B07dsnJa12/VsDfgtsL2mf3ND6BLUL8VqxLQMelbQ58NnCvCtIlaVZkibmY/eqWiuxjvkxcGSlN1bpZhB7N7mOof7nrTKJNMz4UaVrmL/c4PueInViTAR+oXRN9BRJb88V5n/k2H0H4M6ZSPo+fRBA0r8zuMPsVFKl9AAGV/p+Bnwsd5oqlyF7VXWgNbOdM4FPSdpc6WY6h1ZmRMQi4I/AUZLWz3mzlaTX5XUdImm60rXm4yUdSMrRRu4seSrw75J2ULo+7+vA5YWy/iLSUPWbc+f3AGkI/d0RUe/uwsPtq5VbsdPguaTv7yE7q4bofJpPGgJcrBNPiIhLh9j+fFInc/E960VErbrFcB0ovwSm57rVO3DDrfsaqfDmf+DngT+pcJe6gkmka4oeBMZL+hLp2rminST9W640HkL6gq1VqW6q4jFcgdyESaThcn+XtCupodVKk0j7/BBpjPPXKzNyr+TxwHeU7m44TtJuGnyRtg1tKWmM9uVKN5y5jNRDObPB9+9PakwvJA3l+XJEnF+YfxHpzO4VhdeTqNNjWi3S3cpq/iYi6ZrP9+Z9+BlwRp113EtqvB2qJn+IOA/n3Jd0bcZDpJ7kqxj+7rFfIQ0RfYzU+PtNYZ1PA/9KOit/L+ks+nuaicta7rukDqg/SlpK+hy8osl11P2ft9DRpOt/lpBi/EOjb8yV338jjao4nlXXVS8k9Ua/jho397H2yCMQjgL+Rro2fHvSdT2V+ZcDy0k9+L8vTL+KNEzrB6SzW3PJN/wYyXZIZecfgRtIDa7fkeollUb8B4C1SDeLeoQ03LZys48n87rvJ+XkJ0g3gRr2NzIj4gLg/5GGoi0idcAWL9W4lJTrle+Km0kdtXW/OxrYVyu3T0jaIndKfZ70nV63s2qYzqcfA4cr38xO6fKafYfZ/rnACyW9X+lupmtK2kWDh/5WDNmBkjsXBkg3k7o7Im4Z0REpiyjBhXajfZBOt55JOq2/PP/9CanhNYPBF8B+hFV3kFs5j3TB4nGkswaLgM+x+gXsvyIl71JSobpjYb3FZdcAPk26SHkpafjg16tinsbgm4VsQBputoBU2bgW2C/PG7QPUXVhZmHau/K+LSUl/Q+Ak+tsb4DCTR1qbaPGcV4POCev/x7Sl0jxxhITSJWZ+/I+XIwvtPejTY/8OVsIvL7bsfjhhx9+tPpBuvb9nm7H4Ud/PXJ99nBSA/1R0o3/1iV1XAyQGmW3k0bsVG6esympM/ix/J4BYNvCOt9Pur7+cdIZsuMbiGMbUsfbg6TO2j8DO+R5g+rApLulXpm3vYh0lm1S1fYD+Gy3j+9oH8o7ZGZWeko/J3A5qdfvs6Re5efHqhv2mJn1JKWfz3k96azbFNIZsMsiYqTXBZvZGDMmhkqa1SPpeEmLJd1YmLax0m+D3JH/blSYd7jSj5DelhsJVi67kc5gLyENcdwn0h3+fixpWY3Hj7sbrplZw0Qa5vsIadTNLcCXuhqRmZWKz7jZIEo/mPiTGrPuiRp3ryo7Sa8lndb/eUS8JE/7FulawFmSDiPdjOZQSduSfkJhV9KQgD8BL4x0HZSZmZmZtUC+KeDva82LVT/oblXccLMxT9I00i2cKw2320g/HL1I0qbAQERsI+lwgIj4Rl7uPOCIiPhbdyI3MzMzM0tK8dtFkydPjmnTpq02ffny5UycOLHzAQ2hbDGN9XiuvvrqJRHxrJatMJkS6U6e5MZb5ffyNmfwXUIXUOd3RiQdBBwEMGHChJ2mTp262jLPPPMMa6xRjtHIjqW2dsdy++23tyN/W6Ze2dsOZSmrHEdjcbSp7G2pYv6W5XgWlTEm6I+4yp6/tcresv5fWq0f9nM0+zhs7nb77igRwU477RS1XHjhhTWnd1PZYhrr8QBXxejvkDQNuLHw+tGq+Y/kv8cA7ytMP450O+Wez1/HUlu7Y2lF/rbzUS9326Es/3fHMVi9OMqeu1GVv2U5nkVljCmiP+Iqe/7WKnvL+n9ptX7Yz9Hs43C5W45ub7POeiAPkST/XZynL2Dwj07W+4F1MzMzM7OOKsVQyXrm3PcYMw77bVPvmTdrrzZFY2PIbOBAYFb+e05h+qmSvkO6OcnWrPqx6qY1m7/OXSuLaU2Wu+D8tfJoNn+du1YWLnttOKVuuJmNlqTTgOnAZEkLgC+TGmxnSvoQcC+wL0BE3CTpTNKPTq4APhG+o6SZmZmZlYAbbjamRcT+dWbtXmf5I4Ej2xeR2dg0XE/xzO1XDDoD7V5iMzOz5vgaNzMzMzMzs5Jzw83MzMzMDJB0vKTFkm4sTNtY0vmS7sh/NyrMO1zSXEm3SXpLd6K2fuGGm5mZmZlZciKwR9W0w4ALImJr4IL8GknbAvsB2+X3/FDSuM6Fav1m2IabpKmSLpR0i6SbJH0qT3fvg5mZmZmNGRFxMfBw1eS9gZPy85OAfQrTT4+If0TE3cBcYNeOBGp9qZGbk6wAZkbENZImAVdLOh+YQep9mCXpMFLvw6FVvQ+bAX+S9ELfnc/MzMzMetCUiFgEEBGLJD07T98cuKyw3II8bTWSDgIOApgyZQoDAwOD5i9btoyZ2zdfVa5eT9ktW7as52JuVjv3cdiGW07USrIulXQLKSn3Jt1mHVLvwwBwKIXeB+BuSZXeh7+1OngzMzMzsy5RjWlRa8GI+CnwU4Cdd945pk+fPmj+wMAAR12yvOkA5h0wfdhlymRgYIDqfR9r2rmPTf0cgKRpwMuByxll78NwPQ8AUyakW0g3o92t+LL1FDgeMzMzs7Z6QNKmub67KbA4T18ATC0stwWwsOPRWd9ouOEmaT3g18AhEfG4VKuTIS1aY9pqvQ/D9TwAfP+UczhqTnM/Ndfunoey9RQ4HjMzM7O2mg0cCMzKf88pTD9V0ndIlwdtDVzRlQitLzTUKpK0JqnRdkpE/CZPdu+DmZmZmY0Zkk4jXQo0WdIC4MukBtuZkj4E3AvsCxARN0k6E7iZdE+IT/ieDtZOwzbclE6tHQfcEhHfKcxy74OZmZmZjRkRsX+dWbvXWf5I4Mj2RWS2SiNn3F4FvB+YI+m6PO3zuPfBzMzMzMysIxq5q+Ql1L5uDdz7YGZmZmZm1nbD/gC3mZmZmZmZdZcbbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcm64mZmZmZmZlZwbbmZmZmZmZiXXyO+4mZlZiUiaBywFngZWRMTOkjYGzgCmAfOAd0fEI92K0czMzFrLZ9zMzHrT6yNih4jYOb8+DLggIrYGLsivzczMbIxww83MbGzYGzgpPz8J2KeLsZiZmVmLeaikmVnvCeCPkgL4SUT8FJgSEYsAImKRpGfXeqOkg4CDAKZMmcLAwMBqy8zcfkXLA54yYfB6a223E5YtW9a1bTsOMzMbDTfczMx6z6siYmFunJ0v6dZG35gbeT8F2HnnnWP69OmrLTPjsN+2Ks6VZm6/gqPmrPrKmXfA6tvthIGBAWrts+MwM7Oy81BJM7MeExEL89/FwFnArsADkjYFyH8Xdy9CMzMzazU33MzMeoikiZImVZ4DbwZuBGYDB+bFDgTO6U6EZmZm1g4eKmlWEtNGMDxt3qy92hCJldwU4CxJkMrwUyPiD5KuBM6U9CHgXmDfLsZo1jNc9ppZr3DDzcysh0TEXcDLakx/CNi98xGZNce/Q2hmNjIeKmlmZmad5t8hNDNrkhtuZmZm1m3+HUIzs2F4qKSZmZl1Ust/h7D4e3Tt+B3Cao389l1ZfyPPcZn1LjfcrG/5Ogszs65o+e8QFn+Prh2/Q1itkd8hLOtv5Dkus97loZLW73ydhZlZB/l3CM3MRsYNN7PBfJ2FmVmb+HcIzcxGzkMlrZ+1/DqLoikT2n+tRaPXA5Tp2gHHYtD8b2f5d7PGDP8OoVkLuSztL264WT9r+XUWRd8/5RyOmtPej1gj11lAua4dcCxm/cu/Q2hmNnIeKml9y9dZmJmZWaMkzZM0R9J1kq7K0zaWdL6kO/Lfjbodp41dwzbcJB0vabGkGwvT6iappMMlzZV0m6S3tCtws9HwdRZmZmY2Ar6pmXVNI2fcTgT2qJpWM0klbQvsB2yX3/NDSeNaFq1Z60wBLpF0PXAF8NuI+AMwC3iTpDuAN+XXZmZmZrX4pmbWMcNegBMRF0uaVjV5b2B6fn4SMAAcmqefHhH/AO6WNJc0/OxvrQnXrDV8nYWZmY1UIzeEmLn9ipW/KecbQowZbbup2bJly5i5/dPtjB1o/KZm7dIPNwVr5z6O9M4J9ZJ0c+CywnIL8rTVtOuufO1OhrIlnOMxMzMz64i23dRsYGCAoy5Z3spYa2r0pmbt0g83BWvnPrb6lneqMS1qLdiuu/K1OyHLlnCOx8zMzKz9ijc1kzTopmb5RIZvamZtNdK7Sta7894CYGphuS2AhSMPz8zMzMysu3xTMyuDkZ5xqyTpLAYn6WzgVEnfATYDtibd+MHM2qDRH96sXGvh6yzMzMxGxD8eb103bMNN0mmkG5FMlrQA+DKpwbZakkbETZLOBG4GVgCfiIj2X2lpZmZmZtYmvqmZlUEjd5Xcv86smkkaEUcCR44mKDMzMzMzM1tlpNe4mZmZmZmZWYe0+q6SZmZmLdfo9ZxFvqbTzMzGEjfczMzMzEqk2Y4Kd1KY9Qc33MzMzMzaaCRnjM3MqvkaNzMzMzMzs5Jzw83MzMzMzKzkPFTSrI/4ugkzMzOz3uQzbmZmZmZmZiXnhpuZmZmZmVnJueFmZmZmZmZWcr7GzczMxqRa13TO3H4FM+pc6+lrOq1XNXP9cuUz4Hw36z0+42ZmZmZmZlZyPuNmZmZmZmZd0e4fqB9LZ5d9xs3MzMzMzKzk3HAzMzMzMzMrOTfczMzMzMzMSs4NNzMzMzMzs5IbczcnafYCx7F0waKZmZmZma3S7pufQOfaE2Ou4WZmZjYS7vgzMxus1Y2eoX5L04bnoZJmZmZmZmYl5zNuZmZmZjYkn5EeGzoxbNDaxw03MzMzsz7jCrxZ6xQ/T40MBx1px0bfN9yaLbhO3GNimyIxK592fLFXF2julTUzMzMbXtsabpL2AL4LjAOOjYhZ7dqWWSs5d8utlY3JVl0kXabGp/O3czx0rLWcu9bLnL/WCW1puEkaBxwDvAlYAFwpaXZE3NyO7XXSnPsea7qi5y/r3jGWc9fGPuev9SrnrvUy5691SrvOuO0KzI2IuwAknQ7sDTiBreycu9bLnL8l1uw1EJ1QouH/zt0xptkz0iP5TJSoY9z5ax3Rrobb5sD8wusFwCuKC0g6CDgov1wm6bYa65kMLGlLhCP0yRHEpG+2KZikbMeo1fFs2cJ1NWLY3IXey9+R5G27VMfS5s9HU7GM1BD7ULr8bTB3W64sOeg4Bnv9N+vGUbrchSHztxTHs6gs/+NqYymuMVb2lvL/0mplzb9WamQfR5q77Wq4qca0GPQi4qfAT4dciXRVROzcysBGq2wxOZ6WGzZ3offy17HUVqZYWqQlZW87lOVYO45yxsEoy94S7cdKZYwJHFebjLrs7fH9b1g/7Gc797FdP8C9AJhaeL0FsLBN2zJrJeeu9TLnr/Uq5671MuevdUS7Gm5XAltLep6ktYD9gNlt2pZZKzl3rZc5f61XOXetlzl/rSPaMlQyIlZIOhg4j3Rb1OMj4qYRrKrjw3kaULaYHE8LtTB3oVzHwrHUVqZYRq3F+dtqZTnWjmOwUsTRgtwtxX5UKWNM4LharkVlb8/uf5P6YT/bto+KWG0IuZmZmZmZmZVIu4ZKmpmZmZmZWYu44WZmZmZmZlZypW24SdpD0m2S5ko6rI3bmSrpQkm3SLpJ0qfy9CMk3Sfpuvx4a+E9h+e4bpP0lsL0nSTNyfO+J6nW7WEbiWleXs91kq7K0zaWdL6kO/LfjToRj6RtCsfgOkmPSzqkm8en7DqVu4Xt1cvhpnOmhTGNk3StpHO7GYukDSX9StKt+fjs1s3jMtbVy8WqZaZLeqxQdnypTbGsVo5WzVcuh+ZKukHSjm2IoWb5WbVMW46HpOMlLZZ0Y2Fa3dyvem9Hy7DR6FasZSx3q+IrRRlcFZPL4zp66TNXS63ydiT/W5WonthsGdrsPklaW9IZefrlkqY1FFhElO5BurDzTuD5wFrA9cC2bdrWpsCO+fkk4HZgW+AI4DM1lt82x7M28Lwc57g87wpgN9Lvefwe2HOEMc0DJldN+xZwWH5+GPDNTsVT9X+5n/TjgF07PmV+dDJ3G8jhpnOmhTF9GjgVOHek+duiOE4CPpyfrwVs2M3jMtYf9XKxapnplbxocyyrlaNV89+ayyEBrwQub3M8K8vPThwP4LXAjsCNhWk1c79GnB0tw0Z5TLsSaxnL3ar4SlEGV8Xk8rj2cemZz9wQ+7BaeTuS/y0lqic2U4aOZJ+AjwM/zs/3A85oJK6ynnHbFZgbEXdFxD+B04G927GhiFgUEdfk50uBW4DNh3jL3sDpEfGPiLgbmAvsKmlTYP2I+Fuk/8LPgX1aGOrepEKP/HefwvROxbM7cGdE3DNMnN04PmXRsdytGCKHm8qZVsUjaQtgL+DYwuSOxyJpfVLBexxARPwzIh7tRiz9YgTlaTftDfw8ksuADXM51S6NlJ8tExEXAw9XTa6X+0UdL8NGoWuxlq3cLSpLGVwVk8vj+nrpM9eMMtRbR6zJMnQk+1Rc16+A3Rs5w1jWhtvmwPzC6wV04Ms/n6Z8OXB5nnRwHkJzfOF0aL3YNs/Pq6ePRAB/lHS1pIPytCkRsQjSFwbw7A7GU7EfcFrhdbeOT5l1JXcrqnK42ZxplaOBzwHPFKZ1I5bnAw8CJ+QhQ8dKmtilWPpOjfK0aDdJ10v6vaTt2hRCrXK0qNP/7+rys6gTxwPq535RL30OShFrScrdorKUwUUuj+sbC/ta1nprq7Vyn1a+JyJWAI8BmwwXQFkbbrVanG393QJJ6wG/Bg6JiMeBHwFbATsAi4CjhomtlTG/KiJ2BPYEPiHptUOF3oF4UPpBybcDv8yTunl8yqxr+1kjh+suWmNaS2KU9DZgcURc3ehb2hUL6XcqdwR+FBEvB5aThjZ0I5a+MkwuXkMaLvgy4PvA2W0KY7hytGP/7xrlZ1Gnjkejeulz0PVYy1DuVsVTpjK4yOVxfWNhX0tXb+2wkezTiPa3rA23BcDUwustgIXt2pikNUkF7ykR8RuAiHggIp6OiGeAn7HqFH292Bbk56OOOSIW5r+LgbPyth+oDOPJfxd3Kp5sT+CaiHggx9a141NyHc3dilo5TPM50wqvAt4uaR5puMcbJJ3cpVgWAAsionLG51ekikM3YukbdXJxpYh4PCKW5ee/A9aUNLnVcdQpR4s6+f8eVH5WxdmR45HVy/2iXvocdDXWEpW7RWUqg4tcHtfX8/ta0nprO7Ryn1a+R9J4YANWH5q5mrI23K4Etpb0vNxTuR8wux0byuNJjwNuiYjvFKYXr3V4B1C5q8xsYL98N5jnAVsDV+RTpkslvTKv8wPAOSOIZ6KkSZXnwJvztmcDB+bFDiysu63xFOxPYZhPt45PD+hY7lbUy2GazJlWxBIRh0fEFhExjbTvf46I93UplvuB+ZK2yZN2B27uRiz9YohcLC7znMo4fkm7kr6HHmpxHPXK0aLZwAeUvBJ4rDIEpg0GlZ9Vsbb9eBTUy/2ijpdho9C1WMtU7haVqQyuisvlcX299JlbTYnrre3Qyn0qrutdpM/q8GcYowR3o6n1IN3x63bSnVm+0MbtvJp0avIG4Lr8eCvwC2BOnj4b2LTwni/kuG6jcMcbYGdSst4J/ADQCOJ5PunONNcDN1X2nTTu9QLgjvx3407Ek9ezLqkisUFhWleOTy88OpW7DeRw0znT4rims+qOZl2JhTSU96p8bM4GNur2cRnLjyFy8WPAx/IyB+ey7XrgMuBf2hBHvXK0GIeAY/L/ew6wc5uOSa3ys+3Hg9RQXAQ8RerZ/VC93Ac2A35XeG9Hy7BR7mdXYi1ruVsVY9fL4Kp4XB7XPzY985mrEXvp6q0t2q+Gy9CR7BOwDmn4/FxSp8TzG4mr8mYzMzMzMzMrqbIOlTQzMzMzM7PMDTczMzMzM7OSc8PNzMzMzMys5NxwMzMzMzMzKzk33MzMzMzMzErODTczMzMzM7OSc8PNzMzMzMys5MZUw03SiZK+1u04zHpdKz5LkmZIumSI+QOSPjyabZiZlZmkbSRdK2mppE92aJufl3RsJ7ZlvUHSayTdVoI4XE8fpTHVcGs3SRtK+pGk+yU9IWmOpH/v0LanS1rQ4nWurDgruVjSl6qWOVDSnZLWbeW2zczq6UZltx3aUW5bz/kcMBARkyLieyNZQa1OLkl7S7pO0uOScxUqOQAAIABJREFUlki6QNI0gIj4ekS4U8xWioi/RMQ2o12PpJD0glbElMvHZyQty48Fks6UtEsr1j/MtiXps5LukPSkpHslzZK0dru3PVpuuDVI0lrAn4Atgd2ADYDPArMkfbqbsbVCRATwIeDTkrYDkPQs4NvAhyPiiVZsR9L4VqzHzMa0dlV2a31Zfz2X72btsCVwUytXmCvOPwdmkuoizwN+CDzTyu1Y7+jhutXCiFgPmAS8ErgV+Iuk3du83e8BBwEfyNveE3gDcGabtztqPd1wk/RySdfkXtkzgHXy9I0knSvpQUmP5Odb5Hn7Srq6aj0zJZ2dn79V0s15nfdJ+kxe7P3Ac4F9I+LuiHgqIv4AfBL4qqT18/vnSTo8r+MRSSdIWqewrbflXrJHJV0q6aWFefMkfUbSDZIek3RG8b1DHIe9cu/045LmSzqiMG8dSSdLeihv80pJUyQdCbwG+EHu6fhBRNwBHAkcJ2kNUmL/OiIuHCbuw/JZuaV5v99RmDdD0l8l/a+kh4GVsVl51Pss5XkfkTRX0sOSZkvaLE+flnvfxheWra4sS9L3cz7fOlRhLOmDkm7Jn5vzJG3Zjn21ntDyym5W68v6jcDpbdiW9TlJfwZez6rv2U+14rsa2AG4OyIuiGRpRPw6Iu7N6zpC0sn5+Xsk3VWoo+ypNGroWfl1zXJXyf9KWpzL7xskvaSDh8+oX6dUPpsv6VBJ9wMnSFpb0tGSFubH0cpnkFR19l/SZpJ+rVRPvluFkQ2SxikNt63U666WNFXSxXmR63MuvicvP1T9sG7doijn8YKI+BJwLPDNwjq+mz8vj+dYXpOnP0dp9NsmhWV3yvu0pqQXSLoo5++SvH0kbQ18HDggIv4WESsi4ibgncAekt6QlztR0o8lnZ/jv6hYL5H0ojzvYUm3SXp3Yd6Jko6R9Nv83sslbdXkv7+2iOjJB7AWcA/wX8CawLuAp4CvAZvkf8C6pC/nXwJn5/etDTwMvLiwrmuBd+bni4DX5OcbATvm56cDJ9WIYzywAnhLfj0PuBGYCmwM/BX4Wp63I7AYeAUwDjgwL7924b1XAJvl994CfCzPmw4sqHMspgPbkxriLwUeAPbJ8z4K/F8+FuOAnYD187wB0tm04rrGAZcDvwHuzcdvuLj3zTGvAbwHWA5smufNyMfnP/OxmtDt3PGjqc/SG4AlOQfWBr4PXJzfNw0IYHxhXStzqvC/r6z3PcBj/H/27jxerqLO+/jnCwIioBAjkYRIXCIjiKJGxMFxgoggOsZxQRARHJzIDDzomMchiAsujJkZxd1RFExAtrggecSFRa4Mys4AYREJECAkJLInOKAJv+ePqgsnne57u+/t5Zzu7/v16tftrrP07/Strq46VacOTKiz7tuBJcBLcj75BPC7Xn82fvQkP/4aWAc8BqwBPkwqox8B7gaOK6z7dOAHwP3AQ8CVwCTSCajiPr4BTM9pu9W831TgceBv8+v1ysWcjy8pvP5qjuMR4Gry70VetjkwH3gQuIk0KmNZYfnRwD3AauAWYK9ef95+dDw/F8u5mbThtxp4Qc7bXyY1DLesec/jgB8UXp+W8+WzgeXAW3N6w3IX2Cfn760B5XW26/XnOWgPGtQpc15aS2rgbJbLns8ClwHbAs8Bfgd8rpD3luXnG+X/7adIv/8vAG7nqXrsx4DFwI75f/9y4Nl5WQAvKsTXsH7ICHWL2phqjvkNpN7jLfLr9+W8+zRSL/O9wNPzsp8D/1TY9svA1/PzM4Bj8/E+HXhdTj8cuLPB5/0b4Av5+XxSWf36fDxfJf8WAFuQfgc+kON6JamutHNh2weA3fLy04Az25Inep0px5GZX08qgFRI+91whqhZd1fgwcLr/wKOz893Jv3IDjdC7iIVoM+s2ccFwLwGsdxLarmTM+zhhWX7AbcV3vdzNdvewlMVhqXA+wrL/gP49kgZvEE8XwG+nJ//Q/5cXlZnvSFqGm6FzySAWc3EXWf7awvbHgrc1ev84seI+aXhdwk4CfiPQvqWpIJ3Gs013Gr3ewVwcJ11fwEcVlhvI+BPwA69/nz86P6DzlR2R/uxPr7BdoeyfsNtpErEPOC/SRWsqaQK13BlaUfSD/3k/Hoa8MJef9Z+dC8v11k25t9q0rCyhcAfSY24+eQGHBs23LYm1W0WA98ppDcsd0mV5z/k99mo15/joD5oUKfM5eKfh8uevOw2YL/C632Apfn5zEJZ9Bpq6mXAMcD38/NbyHW4OvHUNtwa1g8ZpZ5O44bbX+X3mdIghgeBl+fn7wF+m59vnMvj3fLrU4ATge1rtv8EcFmDfZ8JfDc/n0+hsUWq/6wjle3vAf67ZtvvAJ8ubPu9mv/b79uRJ6o8VHIycE/kTyS7E0DSMyR9R9Kdkh4BLga2lrRxXm8B8F5JIg2BXBgRj+dl7yR9wHfmbtHX5vT7gO1qg1AaJjYxLx92d01Mk/PzHYA5uTv5IUkPkTLA5ML69xae/4mUUUYk6TWSLsrdww+TKigT8+JTgV8BZ+au8/+QtMlI+4vUZQxPDVUaMW5J7y90kz8EvLTw/rWfh5VPw+9SXjb8nIhYQ+rdmNLkvuvtd3Kd9XYAvlrIQw+QzvQ1+z7WpyJiKCIWR8QTEXE96Szq3+bFfyE1ol4UEesi4uqIeKTBriaSRlTUs4J0hrqZeH4QEfdHGl7zJdKZ2OGL/vcnNQAfiIi7SUMzh63L6+4kaZOIWBoRtzXzntYf2vlbHRGXRcT+EfEc0lDK15N6F+qt+xBp5NFLgS8VFjUsdyPi16Se6m8CKyWdqDzc0rquUZ3yjxHxWGHZer/XjPx7O7mmTvdx0mgFSPW7ZsumkeqHI9UtRjKF1HB7CJ68nOnmPOTxIdJ1ncPfm3NIZeoLgL2BhyPiirzsX0n5+QpJN0r6h5xetz6fbUeD+nyu/zyQj2sH4DU1x30Q8NzCti3X55tR5YbbCmBKbnwNe17+O4f0Q/qaiHgmqUCD9A8kIi4jnan4G+C9pAKTvOzKiJhF6mr+KU9dqHgB8GZJW9TE8U7SMJvLCmlTa2Janp/fTfpR37rweEZEnNHaoW/gdGARMDUingV8u3Csf4mIz0TETsBfA28lXd8B6YvRjIZx5/G+3wWOJHWlb006y1z8vzT7PtYbI32XlpMKKABy/n82abjXozm5OONosdCiwX6Xs6G7gQ/V5LHNI+J3rR+O9ZM2VnZH+7H+Y5PxjFSJmMyGlSwAImIJ8BFSb8gqSWcqXy9qA6Mjv9URcSXp8oa616BJ2pXUo3cG659MGLHcjYivRcSrSKNwXkwaQmfd16hOWZsv1vu9ZuTf2ztq/u9bRcR+heXNXo81Ur12pLrFSP4euCYiHs3Xsx1NOim2Ta5jPsxT35vHSPX0g0gdMcX6/L0R8Y8RMZk0OuNbShP7/BqYKmm34ptKmkrqYb6wkDy1sHxL0miK5fm4f1Nz3FtGxD81cXzjUuWG26Wk8b1HSXqapHeQxpJCui7rf4GHJE0APl1n+1NIZ5PWRsQlkGaOlHSQpGdFxF9I1zCsy+ufCiwDfqg0KcMmkvYhFYLHRcTDhX0fIWn7/N4fB87K6d8FDs8VEUnaQmlika2aPWili1KLD+XjfSAiHssZ8b2F9feUtEvubXyEdIZ6+JhWksY2j2akuLcgFR5/zO/3ARr8eFhpjfRdOh34gKRdlS5y/jfg8txb8EdSA+59Shcz/wMbFvbb5v1uIundpOskfl4nhm8Dx+ipGU2fldc3a1dld7Qf69/kpEdpcDJitEoEqaJSW8l6UkScHhGvI1WugsIF+DYQ2vJbLel1SpNGbZtf/xXwNtY/gTy87vB1oB8nXY8zRdI/58UNy11Jr86/+ZuQvhOPFeKx7mpUp6x1BvAJSc+RNJF0DdsP6qx3BfCI0sQmm+ff75fqqWn4vwd8TtL0XOd7mZ6aAKS23jhS/XCkusV68rZTJH0a+GA+TkjfmbWkOubTlG5ZVdvzewppSPvbiserNBnh9vnlg6Qyd11E/IGU90+TtHs+/p2BHwMXRMQFhX3vl79vmwKfI9V/7gZ+BrxY0sG5frNJ/s68pN7xtVNlG24R8WfgHaR/1oOk8aY/yYu/QrpQ8z5SQfbLOrs4ldTAOLUm/WBgqdIQy8NJ1zOQh1K+kdTKvpxUsJ4AHBsR/1mzj9OB80gXe95OulaIiLgK+EdSg/FB0kXBh7Zw2FNIDdLi44Wk2XE+K2k16YtanM70ucCPcrw3kyonwxn7q8C7lGYqajjl9khxR8RNpKEXl5K+0LuQLp61ihjpuxQRFwKfJBVoK0j57YDC5v9IOgt7P+msbG0P2eWkSSHuI00Y8a6IuL9ODGeTKrFn5u/eDaQZ/8zaUtkd5cf6d6RRFZCu0X2H0pD7F5Fuk1KMZaRKxEJSRXibXGH4P4VYd5T0hnwC5DFS+e2K8GBp12/1Q6RK6mJJa0h1nLNJ18XX+gLpOqL/yvWY9wGflzR9lHL3maRK+YOknuP7SbcHsu6rW6es4/PAVcD1pOsZr6m3bkSsA/6OPDsp6ff5e6TRA5Dqtgvzez5CutZ987zsOGCB0vDA/UepH45UTx82OefhNaTJpXYBZkbEeXn5r0jXYv6BlA8fo+bym4j4LWkyk2siYmlh0auBy/P+FwEfjog78rIj8zH/IL/3L0nXkr6zJr7TSZ0/D5CuoT4ov+dq4E2k+tBy0rDI4YliOkrrDz0dHJI2J82E88pI0+C3a79LSRcRXzDaumZmtiFJQ6TJFb4n6V2kk0MTSJXZpcDWEfE+SQeSKhLbk358zwI+GhFrla5PXkC6du3UiDhK6TYnHyNVNKaSZj37EanMfji/90TSj/VrSRWg84E3RsTrcgPxRNJMuo+SZjD757z9BZKeQWocvo30Y/59UmVhe6Upsr9H6nX+C6mxODsi6g1lMjNrW51SaYr770VEM6OsKkfp1hunR8T32rjP+aSTHp9o1z7bYZAbbh8lTYn7hjbvdyluuJmZlZ6kz5KmRH99nsDBzKw02thwO4o0U2Snb2zddXmI5/mk4fSr27jf+ZSw4VbVO62PS/4iiPSDbWZmAygiPiVpFekat3pD6s3MKk3SV0mjAA7pdSztJmkBqS7/4XY22sps1B43SSeTLvheFREvzWkTSENSppGGrewfEQ/mZceQrglYBxwVEb/qVPBmZmZmZmaDoJnJSeYD+9akzQUujIjppGkz5wJI2ol0od7OeZtv6al7p5mZmZmZmdkYjDpUMiIuljStJnkW6Y7nkC7+HiJNkTyLdJfxx4E7JC0hTf156UjvMXHixJg2bf23ePTRR9lii9pbplWTj2Xsrr766vsi3WC0tOrlX6je/93xtl/Z82+V865jbI9GMZY970K182+7DeIxQ3Xzb6O8WyWDlue6dbyj5d2xXuM2KSJWAETEiuF7iZCmqy/eR2RZTtuApNnAbIBJkybxxS+uP8vsmjVr2HLLttxkvOd8LGO355573jn6Wr01bdo0rrrqqg3Sh4aGmDlzZvcDGiPH236SSp1/q5x3HWN7NIqx7HkXqp1/220Qjxmqm38b5d0qGbQ8163jHS3vtntyEtVJq3sRXUScSJpWmRkzZkTth9FPGcLHYmZmZmZm4zHWG3CvlLQdQP67KqcvI90bZ9j2pHvZmJmZmZmZ2RiNtcdtEWla0Xn57zmF9NMlnQBMBqYDV4w3yKpbfM/DHDr33KbXXzrvLR2MxrrF/3erqlbzLjj/Wnm47DVrn2kNvktzdlnb8Hvm71TnjNpwk3QGaSKSiZKWAZ8mNdgWSjoMuAt4N0BE3ChpIXATsBY4IiLWdSh2MzMzMzOzgdDMrJIHNlhU9+7rEXE8cPx4gjIzMzMzM7OnjPUaNzMzMzMzM+sSN9zMzMzMzMxKrt23AzAzMzMzswHVaEKTRjyZSfPc42ZmZmZmZlZybriZmZWQpKmSLpJ0s6QbJX04p0+QdL6kW/PfbQrbHCNpiaRbJO3Tu+jNzMys3dxwMzMrp7XAnIh4CbA7cISknYC5wIURMR24ML8mLzsA2BnYF/iWpI17ErmZmZm1nRtuZmYlFBErIuKa/Hw1cDMwBZgFLMirLQDenp/PAs6MiMcj4g5gCbBbd6M2MzOzTvHkJGZmJSdpGvAK4HJgUkSsgNS4k7RtXm0KcFlhs2U5rXZfs4HZAJMmTWJoaGiD95u0OczZZW1LMdbbTyetWbOm6+/ZKsdoZmbt5IabmVmJSdoS+DHwkYh4RFLDVeukxQYJEScCJwLMmDEjZs6cucFGXz/tHL60uLWfh6UHbbifThoaGqJe7GXiGM3MrJ08VNLMrKQkbUJqtJ0WET/JySslbZeXbwesyunLgKmFzbcHlncrVjMzM+ssN9zMzEpIqWvtJODmiDihsGgRcEh+fghwTiH9AEmbSXo+MB24olvxmpmZWWe54WZmVk57AAcDb5B0bX7sB8wD9pZ0K7B3fk1E3AgsBG4CfgkcERHrehO6DTLfysLMrDN8jZuZWQlFxCXUv24NYK8G2xwPHN+xoMyaM3wri2skbQVcLel84FDSrSzmSZpLupXF0TW3spgMXCDpxT7xYGa2Pve4mZmZWdv4VhbWryQtlbQ4j4C4Kqc17Ek2azf3uJmZmVlHtPNWFnl/bb+dRT/cDmFQb+vQo+PeMyLuK7yeS52e5G4HZYPBDTczMzNru3bfygI6czuLbt/KohMG9bYOJTnuWcBwEAuAIdxwa8m0uee2tP7SeW/pUCTl54abmZmZtdVIt7LIvW2+lYVVUQDnSQrgO/lEQqOe5PU001tcRo16r1vt2W6nXnx2ZenVdsPNzMzM2qaJW1nMY8NbWZwu6QTS5CS+lYWV1R4RsTw3zs6X9PtmN2ymt7iMDm3QGzZnl7Ut9Wy3Uy96yUvSu+uGm5mZmbXV8K0sFku6Nqd9nNRgWyjpMOAu4N2QbmUhafhWFmvxrSyspCJief67StLZpEl0GvUkm7WdZ5W0vibpZEmrJN1QSPO9hMzMOiQiLokIRcTLImLX/Ph5RNwfEXtFxPT894HCNsdHxAsjYseI+EUv4zerR9IW+fYWSNoCeBNwA0/1JMP6PclmbeeGm/W7+cC+NWnDM0BNBy7Mr6m5l9C+wLckbdy9UM3MzKykJgGXSLqONJT33Ij4JakneW9JtwJ759dmHeGhktbXIuLiPB11UaMZoJ68lxBwh6Thewld2o1YzczMrJwi4nbg5XXS7wf26n5ENojccLNB5HsJ1VGWGZOaVbV4zczMzMbDDTezpwz0vYTKMmNSs6oWr5mZmdl4+Bo3G0Qr88xP+F5CZmZmZlYFbrjZIGo0A9Qi4ABJm0l6Pr6XkJmZmZmVhIdKWl+TdAZpIpKJkpYBn8b3EjIzMzOzinHDzfpaRBzYYFHdGaAi4njg+M5FZGZmZmbWOg+VNDMzMzMzK7lx9bhJWgqsBtYBayNihqQJwFnANGApsH9EPDi+MM3MzMzMzAZXO3rc9oyIXSNiRn49F7gwIqYDF+bXZmZmZmZmNkadGCo5C1iQny8A3t6B9zAzMzMzMxsY452cJIDzJAXwnXxT4kkRsQIgIlZI2rbehpJmA7MBJk2axNDQ0HrL16xZs0FaVU3aHObssrbp9ct83P30fzEzMzMzq4rxNtz2iIjluXF2vqTfN7thbuSdCDBjxoyYOXPmesuHhoaoTauqr592Dl9a3PxHvfSgmZ0LZpz66f9iZmZmZlYV4xoqGRHL899VwNnAbsBKSdsB5L+rxhukmZmZmZnZIBtzw03SFpK2Gn4OvAm4AVgEHJJXOwQ4Z7xBmpmZmZmZDbLx9LhNAi6RdB1wBXBuRPwSmAfsLelWYO/82szMWiDpZEmrJN1QSJsg6XxJt+a/2xSWHSNpiaRbJO3Tm6jNzMysU8Z8jVtE3A68vE76/cBe4wnKzMyYD3wDOKWQNny7lXmS5ubXR0vaCTgA2BmYDFwg6cURsa7LMZuZmVmHjHdykoEzbe65LW8zZ5fOvsfSeW9p7Q3MrPQi4mJJ02qSZwEz8/MFwBBwdE4/MyIeB+6QtIR0zfGl3YjVzMzMOs8NNzOz6mh0u5UpwGWF9ZbltA2MdisWaP0WJtD925hU4dYkgxyjpJOBtwKrIuKlOW0CcBYwDVgK7B8RD+ZlxwCHAeuAoyLiV20Pysys4txwMzOrPtVJi3orjnYrFmj9FibQ/duYVOHWJAMe43w81NfMrK3GdTsAMzPrqka3W1kGTC2stz2wvMuxmT0pIi4GHqhJnkUa4kv++/ZC+pkR8XhE3AEMD/U1Kw1JUyVdJOlmSTdK+nBOP07SPZKuzY/9eh2r9S/3uJmVxFiun/T1jQNn+HYr81j/diuLgNMlnUDqsZhOmu3XrExKOdS37MNZm1GFYbmd0OXjXgvMiYhr8u2wrpZ0fl725Yj4YrcCscHlhpuZWQlJOoM0EclEScuAT5MabAslHQbcBbwbICJulLQQuIlUuTjCw8ysQno61Lfbw3w7oQrDcjuhm8edTzoMn3hYLelmGpxgMOsUN9zMzEooIg5ssKju7VYi4njg+M5FZDZuKyVtl3vbPNTXKivP+PsK4HJgD+BISe8HriL1yj1YZ5tRe4vLqFHv9VgmsWqXXnx2ZenVdsPNzMzMusFDfa3yJG0J/Bj4SEQ8Ium/gM+Reok/B3wJ+Ifa7ZrpLe6G1i/LqN9UmLPL2pYnsWqXXvSSl6VX2w03MzMzaysP9bV+JGkTUqPttIj4CUBErCws/y7wsx6FZwPADTczMzNrKw/1tX4jScBJwM0RcUIhfbvhSXeAvwdu6EV8NhjccDMzMzMzG9kewMHAYknX5rSPAwdK2pU0VHIp8KHehGeDwA03swprday6bx9gZmbWuoi4hPozoP6827HY4PINuM3MzMzMzErODTczMzMzM7OSc8PNzMzMzMys5HyNWx/wdU5mZmZmZv3NPW5mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZXcwE9O0urEHmZmZmZm1huDPCmfe9zMzMzMzMxKzg03MzMzMzOzkhv4oZJmZmZmZlXkS34GixtuZgNkpAJ+zi5rObRmeT+NCzczMzOrMjfczMxs3Ab5YnEzM7NucMPNzMzMzMz60liGk5b15KInJzEzMzMzMys597gNoH4682Bm1eShlWZm6/NEIzaavmu4OdObmZnZIFl8z8MbTC41Gp8MMauejg2VlLSvpFskLZE0t1PvY9ZuzrtWZc6/VlXOu1Zlzr/WDR3pcZO0MfBNYG9gGXClpEURcVMr+3HvWXVVdRhUu/Juv6jq/3FQOf9aVTnv2ljqfPP33aIDkbTO+bf/1ObHerdMKupW/adTQyV3A5ZExO0Aks4EZgHOwBU1nIFHy7h9wHm3y8Z6gqbZvNhqYVrxa0D7Nv+2+iM6FiX6Pw6ivs27NhCcfwdct050d6rhNgW4u/B6GfCa4gqSZgOz88s1km6p2cdE4L4OxddVR/lYRqV/b7hoh3a/1yhGzbvQVP6Fiv3f2/G/HeH/2HbNxtuNmKqUf/sl73aiLOpAXin950jjGEuXd6Ez+beb5VYHtZzX+uG49/z36uTfJvNuZfRT3bYZ7T7esdYbOtVwU520WO9FxInAiQ13IF0VETPaHVgv+FgqZdS8C6PnX6jeZ+V4+8K4y16oxmfrGNujRDEObNnbDoN4zFCq425L2VslJfrsu6Isx9upyUmWAVMLr7cHlnfovczayXnXqsz516rKedeqzPnXuqJTDbcrgemSni9pU+AAYFGH3susnZx3rcqcf62qnHetypx/rSs6MlQyItZKOhL4FbAxcHJE3NjibvqmOxkfS2W0Ke8Oq9pn5Xgrro35twqfrWNsj1LEOOBlbzsM4jFDSY67zfm3Kkrx2XdRKY5XERsMITczMzMzM7MS6dgNuM3MzMzMzKw93HAzMzMzMzMrudI13CTtK+kWSUskze11PM2QdLKkVZJuKKRNkHS+pFvz320Ky47Jx3eLpH16E/WGJE2VdJGkmyXdKOnDOb1yx9JLVcvDkpZKWizpWklX9TqeWq1+v2x8qpB/y5hnq5BPG8R4nKR78md5raT9ehnjeFQh77Zbvf9pv2tUV7H26Zd6bbOqVP8tVcNN0sbAN4E3AzsBB0raqbdRNWU+sG9N2lzgwoiYDlyYX5OP5wBg57zNt/Jxl8FaYE5EvATYHTgix1vFY+mJCufhPSNi1zLco6SO+TT5/bLxqVj+LVuenU/58+l8NowR4Mv5s9w1In7e5ZjaomJ5t53mU/9/2s8a1VWsfebTH/XaZlWm/luqhhuwG7AkIm6PiD8DZwKzehzTqCLiYuCBmuRZwIL8fAHw9kL6mRHxeETcASwhHXfPRcSKiLgmP18N3AxMoYLH0kOVzMNl1uL3y8bH+XeMqpBPG8TYLwYy7/b5/7SuEeoq1ib9Uq9tVpXqv2VruE0B7i68XkZ1v4yTImIFpAwBbJvTK3GMkqYBrwAup+LH0mVV/EwCOE/S1ZJm9zqYJjXKkzY+Vcm/VcmzVcmnR0q6Pg+Pquqw46rkXWujmrqKddZA1AXLXv8tW8NNddL67X4FpT9GSVsCPwY+EhGPjLRqnbRSHUsPVPEz2SMiXkkaYnSEpNf3OiDrmarkX+fZ9vkv4IXArsAK4Eu9DWfMqpJ3rU1aqKtYZ/XNd68K9d+yNdyWAVMLr7cHlvcolvFaKWk7gPx3VU4v9TFK2oSUaU+LiJ/k5EoeS49U7jOJiOX57yrgbKoxxKFRnrTxqUT+rVCeLX0+jYiVEbEuIp4Avkt5P8vRVCLvWns0qKtYZ/V1XbAq9d+yNdyuBKZLer6kTUkX/i3qcUxjtQg4JD8/BDinkH6ApM0kPR+YDlzRg/g2IEnAScDNEXFCYVHljqWHKpWHJW0haavh58CbgCrMTtYoT9r4lD7/Vizu/a/nAAAgAElEQVTPlj6fDldKsr+nvJ/laEqfd609RqirWGf1bV2wUvXfiCjVA9gP+ANwG3Bsr+NpMuYzSENM/kJqhR8GPJs0A82t+e+EwvrH5uO7BXhzr+MvxPU6Ulfv9cC1+bFfFY+lx59jZfIw8ALguvy4sYzxtvr98mPcn3ep829Z82wV8mmDGE8FFudyfxGwXa8/y3EcX6nzbrf+p72OqQvHXLeu0uu4+unRL/XaFo63MvVf5Tc3MzMzMzOzkirbUEkzMzMzMzOr4YabmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuPUBSUslvbHXcZiZVY2kHSX9j6TVko7qdTxmZmaNuOFmZm0j6SBJ53Vgvz45YZ3yr8BQRGwVEV/rdTBmRZLmS/r8OLYfkvTBMW77bUmfHOt722Dw73N3ueFmZm0TEadFxJuGX0sKSS/qZUxm9Uh6Wn66A3BjL2Oxamul4lqlSm5EHB4Rn+t1HFZdhXLW2sQNtw6S9ApJ1+QhOGdJOlPS5yUdKumSmnWfrOBKekseuvOIpLslHVez7sGS7pR0v6Rju3hIZmalIOloSffk8vUWSXvV9k5ImilpWeH10rzd9cCjkn4N7Al8Q9IaSS9uovx9naTfSXooLz80p28m6YuS7pK0MvdWbN6VD8Osy1whNwBJpwLPA/5fLkP/NddnD5N0F/DrvN4PJd0r6WFJF0vaubCP+ZK+JekXeR+/lfRcSV+R9KCk30t6RWH9Dcr+rh94D7nh1iGSNgV+CpwKTAB+CLyzyc0fBd4PbA28BfgnSW/P+90J+C/gYGAy8Gxg+7YGb5UxjsrrxyRdL+lRSSdJmpQLzdWSLpC0TV53Wi6EP5ArqQ9KOlzSq/P2D0n6RmHfT56UkHRxTr4uF8bvGeVY3irp2rzP30l6WYP1NpI0V9Jt+eTFQkkT8rJfSjqyZv3rJL2jxY/WSkzSjsCRwKsjYitgH2Bpk5sfSCpXt46INwD/DRwZEVtGxB8Yufx9HvAL4OvAc4BdgWvzfv8deHFOexEwBfjU+I7Uyq5BxfVtkm7MZdmQpJc0WjenN6zUthDHrFx+PpLLxn0Li3fIleHVks6TNLGw3WgV6s/n5zMlLcu/OfcC35c0UdLP8nE+IOm/JbleOUAi4mDgLuDvImJLYGFe9LfAS0hlM6RyczqwLXANcFrNrvYHPgFMBB4HLs3rTQR+BJwA4y77+4K/YJ2zO7AJ8JWI+EtE/Ai4spkNI2IoIhZHxBMRcT1wBulLAPAu4GcRcXFEPA58EniiA/FbyY2zAHsnsDepovl3pEL146RCciOgdpKG15AK3fcAXwGOBd4I7AzsL+lva9YnIl6fn748V4rPGuFYXgmcDHyIdDLiO8AiSZvVWf0o4O2k78Rk4EHgm3nZ6aSK+fB+dyINhTu30XtbJa0DNgN2krRJRCyNiNua3PZrEXF3RPxvvYWjlL8HARdExBm5XL8/Iq6VJOAfgX+JiAciYjXwb8AB4zpKK706FdefkvLMR0iN+5+TGmqb1q4bEf+RdzNapXZEknYDTgE+Rjrh8HrW/y14L/CBvP9Ngf9bWNbKez+XdCJ6B2A2MAdYlo9zEuk3JFqJ3frWcRHx6HA5GxEnR8TqXG89Dni5pGcV1j87Iq6OiMeAs4HHIuKUiFgHnAUM97iNp+zvC264dc5k4J6IKBZidzazoaTXSLpI0h8lPQwcTqpQD+/37uF1I+JR4P42xWzVMp4C7OsRsTIi7iH1OFweEf+TC9WzeaqQHPa5iHgsIs4j9UicERGrCtvXrt+qfwS+ExGXR8S6iFhAOuu2e511PwQcGxHLCj8C71IaunM2sKukHfK6BwE/yetZn4iIJaSK8XHAKqVh6JOb3PzukRaOUv5OBep9x54DPAO4Ovc+PAT8MqfbYHkPcG5EnB8RfwG+CGwO/HWjDZqo1I7mMODk/J5PRMQ9EfH7wvLvR8QfciV6IalXeCzv/QTw6Yh4PO/rL8B2wA75RMZ/19R5bHA9Wc5K2ljSvNwT/AhPnVSYWFh/ZeH5/9Z5vSWMu+zvC264dc4KYEo+Ezvsefnvo6QfeQAkPbdm29OBRcDUiHgW8G1geD8rSJWH4W2fQeqhsAEzzgKsqUJyHOu3agdgznClN1d8p5JOVNRb9+zCejeTGrGTck/HuTzV03EALZ69tmqIiNMj4nWk/BCkoYrrla2kHoINNh1l1yOVv3cDL6yzzX2k78HOEbF1fjwr98DYYJlM4SRtRDxByjdT6q3cZKV2NI1OKAy7t/D8T+Tyegzv/cfcIzLsP4ElwHmSbpc0t4WYrX/UK1OLae8FZpFG6TwLmJbTxRg0KPsHhhtunXMpsBY4StLT8jU2u+Vl1wE7S9pV0tNJFe+irYAHIuKxPATivYVlPwLeqnSB/KbAZ/H/cWCNo/JaNncDxxcqvVtHxDMi4owG6765Zt2n594/SMOUDpT0WtKZ7ou6dAzWJUr3XntDHkr7GKnRtI50vdl+kibkE2IfGcPuRyp/TwPeKGn/XK4/W9KuuXL+XeDLkrbNMU6RtM+Gu7c+VKykLieVxwDkk7dTgXvqrAvtqdQ2OqEwmlbfe73Yc0/dnIh4AWnI/Uc1YBNFGJBO5L5ghOVbkUbQ3E+qm/zbWN9ohLJ/YLjC3yER8WfgHcChpGtw3gP8JC/7A6nBdQFwK3BJzeb/DHxW0mrSxe3DF3sSETcCR5DOCq/I+16GDZwOV17bZbQCfdh3gcPzMDVJ2kJpdr+t6qz7beD44eGQkp4jaVZh+c9JFafPAmflSrX1l82AeaSerntJ1+d8nDQZ1HWknoPzSNdGtGqk8vcuYD/StT0PkL5rL8+Ljyb1PlyWey8uAHYcw/tb9RTLuYXAW5QmitqElFceB35XZ11oT6X2JOAD+T03yicN/qqJ7cb13koTSr0oN04fIf3+DFQl2gD4AvCJPALmXXWWn0Lqhb4HuAm4bBzv1ajsHxwR4UeXHsB84PO9jsOP/ngALwOuAFaTKpE/Iw3TeTqpwvoIcD3wL8CywnZLgTcWXv+AdCHx8OsPkiZggHQGNoCnFZYvA2bWbP+J/PxQ4JLCssNJJxgeAvYf5Xj2JU3g81De5ofAVrUxk044fRS4JR/7bcC/1ezrpBz3q3v9f/LDDz/6+0Hqtborl13/F/h7UgX1YeA3pCG0jdbdEjgnl2V3kmY0DeBFef2m6g35Pa/P+1kC7JPTh4APFtZ7soxu5b2BmcXfkZz2L7lsfjT/Lnyy1/8LP/zo94cifB1pt0iaTyr4PtHrWMzMzMzMrDo8VNL6mqSTJa2SdEMhbYKk8yXdmv9uU1h2jKQlSvdE8/UpZmZmZlYKbrh1UUQc6t62rptPGoJXNBe4MCKmAxfm18P3/DqAdG+yfYFvSdq4e6H2N0kfV7rpbO3jF72OzcyszFx+mhngoZLW/yRNI920/KX59S2ka7RWSNoOGIqIHSUdAxARX8jr/Yp07delvYnczMzMzCx5Wq8DAJg4cWJMmzZtg/RHH32ULbbYovsBNaGssZU1LhhbbFdfffV9EdHum9hOiogVALnxtm1On8L6sx0to/G9d2YDswE233zzV02dOpUnnniCjTaqZie2Y++MP/zhD53Iv21ThbK3LLEMWhwdKnvbqgr5t1sG8Zih8XGXPf9WOe86xvYYc97t9ewoEcGrXvWqqOeiiy6qm14GZY2trHFFjC024KoYZ/4izYx4Q+H1QzXLH8x/vwm8r5B+EvDO0fY/nH/L/NmPxrF3RjvybycfVSh7yxLLoMVR9rwbFcm/3TKIxxzR+LjLnn+rnHcdY3uMNe+W8zS1WWetzEMkyX9X5fRlpBulDtuedDNVMzMzM7OeKsVQyUYW3/Mwh849t6Vtls57S4eisT6yCDiEdBPHQ0j3sRlOP13SCaT7oU0n3SdtTKY575o11Oz3Y84uazl07rn+fljHtFpWg8tr6wzXe200pW64mY2XpDNINw6dKGkZ8GlSg22hpMNIN0J9N0BE3ChpIenGqWuBIyJiXU8CNzMzMzMrcMPN+lpEHNhg0V4N1j8eOL5zEZmZmZmZtc7XuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmZmVnJuuJmZmZmZmZWcG25mZmZmZmYl54abmZmZmZlZybnhZmZmZmYGSJoq6SJJN0u6UdKHc/oESedLujX/3aawzTGSlki6RdI+vYve+p0bbmZmZmZmyVpgTkS8BNgdOELSTsBc4MKImA5cmF+Tlx0A7AzsC3xL0sY9idz6nhtuZmZm1laSTpa0StINhTT3WFjpRcSKiLgmP18N3AxMAWYBC/JqC4C35+ezgDMj4vGIuANYAuzW3ahtUDyt1wGYmZlZ35kPfAM4pZA23GMxT9Lc/Promh6LycAFkl4cEeu6HLPZeiRNA14BXA5MiogVkBp3krbNq00BLitstiyn1e5rNjAbYNKkSQwNDW3wfpM2hzm7rG0pxnr76aQ1a9Z0/T1b1c8xuuFmZmZmbRURF+dKb9EsYGZ+vgAYAo6m0GMB3CFpuMfi0m7EalaPpC2BHwMfiYhHJDVctU5abJAQcSJwIsCMGTNi5syZG2z09dPO4UuLW6uaLz1ow/100tDQEPViL5N+jnHU3CHpZOCtwKqIeGlOmwCcBUwDlgL7R8SDedkxwGHAOuCoiPhVy1GZmZlZvxlXjwU012vR6pnsVns4oPu9HKOpQg9DJ3TquCVtQmq0nRYRP8nJKyVtl/PudsCqnL4MmFrYfHtgeduDMqO5Hrf5eLiDmZmNYNrcc3sdglVXUz0W0FyvRatnsg8dQ97tdi/HaKrQw9AJnThupa61k4CbI+KEwqJFwCHAvPz3nEL66ZJOINV9pwNXtDUos2zUyUki4mLggZpkX6BpZmZmrViZeypwj4WV2B7AwcAbJF2bH/uRGmx7S7oV2Du/JiJuBBYCNwG/BI5wh4V1ylivcevKcIcyX6RZ1mEJZY0Lyh2bmZl1nHssrPQi4hLq9wID7NVgm+OB4zsWlFnW7slJ2jrcocwXaZZ1WEJZ44Jyx2ZmZu0j6QzSRCQTJS0DPk1qsC2UdBhwF/BuSD0WkoZ7LNbiHgszs7rG2nDzBZpmZmZWV0Qc2GCReyzMzMZorDfgHh7uABsOdzhA0maSno+HO5iZmZmZmY1bM7cD8HAHMzMzMzOzHhq14ebhDmZmZmZmZr3V7slJzMyswyQtBVYD64C1ETFD0gTgLGAasBTYPyIe7FWMZmZm1l5jvcbNzMx6a8+I2DUiZuTXc4ELI2I6cGF+bWZmZn3CPW5mZv1hFul6ZIAFwBBwdK+Cabdpc89teZul897SgUjMzMx6ww03M7PqCeA8SQF8J98Xc1JErADIt2rZtt6GkmYDswEmTZrE0NDQBuusWbOmbvpI5uyytqX1mzVp87Hvu9VjGMlYPpNOKEscZmbWfW64mZlVzx4RsTw3zs6X9PtmN8yNvBMBZsyYETNnztxgnaGhIeqlj+TQMfSINWPOLmv50uKx/VQtPWhm2+IYy2fSCWWJY9C12gPs3l8zawdf42YDS9JSSYslXSvpqpw2QdL5km7Nf7fpdZxmtSJief67Cjgb2A1YKWk7gPx3Ve8iNDMzs3Zzw80GnSd4sEqRtIWkrYafA28CbgAWAYfk1Q4BzulNhGZmZtYJHipptr6+nuDB+sIk4GxJkMrw0yPil5KuBBZKOgy4C3h3D2M0MzOzNnPDzQZZWyd4qJ00oNUJFb5+WusdJLtMeVbL29RT5QkPqhz7WETE7cDL66TfD+zV/YjMzMysG9xws0HW1gkeaicN6NRkDUXtmnyhyhMeVDl2MzMzs2b5GjcbWJ7gwczMzMyqou963DxFrzUjT+qwUUSsLkzw8FmemuBhHp7gwczMzMxKou8abmZN8gQPZmZmZlYZbrjZQPIED2ZmZmZWJb7GzczMzMzMrOTccDMzMzMzAySdLGmVpBsKaRMknS/p1vx3m8KyYyQtkXSLpH16E7UNCjfczMzMrGskLZW0WNK1kq7KaQ0rxmZdNh/YtyZtLnBhREwHLsyvkbQTcACwc97mW5I27l6oNmjG1XBz4WtmZmZjsGdE7BoRM/LruhVjs26LiIuBB2qSZwEL8vMFwNsL6WdGxOMRcQewhHRrIbOOaMfkJHtGxH2F18OF7zxJc/Pro9vwPmZmZtafZgEz8/MFwBCuO1h5TIqIFQARsULStjl9CnBZYb1lOW0DkmYDswEmTZrE0NDQhm+yOczZZW1LgdXbTyetWbOm6+/Zqn6OsROzSrrwNTMzs0YCOE9SAN+JiBNpXDFeTzOV31YrRK1WlMei05XIKlRUO6EEx606aVFvxZzPTwSYMWNGzJw5c4N1vn7aOXxpcWtV86UHbbifThoaGqJe7GXSzzGOt+HW0cJ3LGceWjXWL3wJCou6yhoXlDs2MzPrmj0iYnmuH5wv6ffNbthM5bfVCtGhc89tet2x6nTlugoV1U7o4nGvlLRdrtduB6zK6cuAqYX1tgeWdyMgG0zjbbh1tPAdy5mHVo21MC1rIVnWuKDcsVXVtBYrHEvnvaVDkZiZNScilue/qySdTbomqFHF2KwMFgGHAPPy33MK6adLOgGYDEwHruhJhDYQxjU5SbHwBdYrfAFc+JqZmdkwSVtI2mr4OfAm4AaeqhjD+hVjs66SdAZwKbCjpGWSDiM12PaWdCuwd35NRNwILARuAn4JHBER63oTuQ2CMXdn5QJ3o4hYXSh8P0vjsxJmZmY22CYBZ0uCVAc5PSJ+KelKYGGuJN8FvLuHMdoAi4gDGyzaq8H6xwPHdy4is6eMZxyiC18zW0+rQzfBwzfNBklE3A68vE76/TSoGJuZWTLmhlu/FL6+RsjMzMzMzMquszN/mJlZ5Sy+5+GuzLRnZmZmzXPDzczM+pJHVJiZWT8Z16ySZmZmZmZm1nnucTMzM7O+4aG+Ztav3ONmZmZmZmZWcu5xMzMzY+Rr4ubssnaDXhxfE2fN8vWWZtYO7nEzMzMzMzMrOfe4mVlP+Uy0mZmZ2ejc42ZmZmZmZlZybriZmZmZmZmVnIdKtmh4WFe9C9Ub8dAuq6pWhzF2Q21Mo30X/f0zMzOzfuCGm5n1NV9DZ2ZmZv3AQyXNzMzMzMxKzj1uZgOkUe9TK0N/zSxxb66ZmXWTe9zMzMzMzMxKzj1uZmYFY5mQxT0pZmbWj1qdEAz8m9hJ7nEzMzMzMzMruY71uEnaF/gqsDHwvYiY16n3Mmsn512rMuff8nJv7sicd63KnH+tGzrScJO0MfBNYG9gGXClpEURcVMn3q/fjOfeWe7CHh/nXasy51+r6oQpzrvra/X/OH/fLToUiTXD+de6pVM9brsBSyLidgBJZwKzAGfgAVHvR2ekRmVZKg8471q1Of/2mbFcX1JRzrvjsPiehzueL0r0O11Gzr8DrlsnWxQRY9pwxJ1K7wL2jYgP5tcHA6+JiCML68wGZueXOwK31NnVROC+tgfYHmWNraxxwdhi2yEintOJYOppJu/m9Hr5t8yf/Wgce2eULv9WsOwtSyyDFkfp8m5Or1r+7ZZBPGZofNyly799lHcdY3uMKe92qsdNddLWayFGxInAiSPuRLoqIma0M7B2KWtsZY0Lyh1bwah5F+rn34ocX12OvW/0XdlbllgcR8eNuezdYEf9+xk1NIjHDKU67r4rextxjO0x1hg7NavkMmBq4fX2wPIOvZdZOznvWpU5/1pVOe9alTn/Wld0quF2JTBd0vMlbQocACzq0HuZtZPzrlWZ869VlfOuVZnzr3VFR4ZKRsRaSUcCvyJNi3pyRNw4hl2N2KXcY2WNraxxQbljA8add0t/fCNw7H2gT8vessTiODqojXkX+vQzGsUgHjOU5Lj7tOxtxDG2x5hi7MjkJGZmZmZmZtY+nRoqaWZmZmZmZm3ihpuZmZmZmVnJlaLhJmlfSbdIWiJpbp3lkvS1vPx6Sa/sUlxTJV0k6WZJN0r6cJ11Zkp6WNK1+fGpLsW2VNLi/J5X1Vne9c9M0o6Fz+FaSY9I+kjNOj35vDpptPzbK/XyiKQJks6XdGv+u01h/WPyMdwiaZ9C+qvyfpbkPFVv2uPxxnqypFWSbiiktS1WSZtJOiunXy5pWruPoYrKUPaWqZwtS7k6qGXpeJW1LG63Rt+ZkcrMfiFpY0n/I+ln+XXfHHPZ82+93+myaeb3pNckPV3SFZKuyzF+pqUdRERPH6SLOG8DXgBsClwH7FSzzn7AL0j3ydgduLxLsW0HvDI/3wr4Q53YZgI/68HnthSYOMLynnxmNf/Xe0k3Euz559Xh4xwx//Ywtg3yCPAfwNz8fC7w7/n5Tjn2zYDn52PaOC+7Anhtzku/AN7cgVhfD7wSuKETsQL/DHw7Pz8AOKvX/59eP8pS9papnC1juTooZWmbPqdSlsUdONa635lGZWY/PYCPAqcP5/1+OeYq5N96v9NlezTze9LrR/792DI/3wS4HNi92e3L0OO2G7AkIm6PiD8DZwKzataZBZwSyWXA1pK263RgEbEiIq7Jz1cDNwNTOv2+bdKTz6xgL+C2iLizi+/ZC83k3zKZBSzIzxcAby+knxkRj0fEHcASYLecZ54ZEZdGKmVOKWzTNhFxMfBAB2Mt7utHwF6d6DmsmFKUvRUrZ3tRrg5KWTpeVSuLx2yE70yjMrMvSNoeeAvwvUJyvxxz6fNvg9/pUqnC70n+/ViTX26SH03PFFmGhtsU4O7C62Vs+CE3s05H5aFVryC1jGu9Nnd5/kLSzl0KKYDzJF0taXad5b3+zA4AzmiwrBefV6f0+nMeSb08MikiVkAq4IBtc3qj45iSn9emd0M7Y31ym4hYCzwMPLtjkVdD6creEpSzZSxXB6UsHa8yl8UdU/OdaVRm9ouvAP8KPFFI65djHsj820mj/J70VB7yey2wCjg/IpqOsSP3cWtRvbPetS3PZtbpGElbAj8GPhIRj9QsvoY0hGWNpP2AnwLTuxDWHhGxXNK2wPmSfp/PhjwZdp1tuvKZKd188m3AMXUW9+rz6pSe5s1RbJBHRli30XGU8fjGEmsZj6PXSlX2lqScLVW5OmBl6XgN3He89jvTz4MIJL0VWBURV0ua2et4OmDg8m8njfJ70nMRsQ7YVdLWwNmSXhoRTV07WIYet2XA1MLr7YHlY1inIyRtQvrnnxYRP6ldHhGPDHd5RsTPgU0kTex0XBGxPP9dBZxN6mYv6tlnBrwZuCYiVtYu6NXn1UG9/JxH1CCPrBwe2pX/rsqrNzqOZfl5bXo3tDPWJ7eR9DTgWZR8yEcXlKbsLUs5W8JydZDK0vEqbVncCQ2+M43KzH6wB/A2SUtJwwjfIOkH9M8xD1T+7aTRfk/KJCIeAoaAfZvdpgwNtyuB6ZKen88uHgAsqllnEfB+JbsDDw93jXdSvgbmJODmiDihwTrPHb5WRtJupM/0/g7HtYWkrYafA28CalvqPfnMsgNpMLSnF59XhzWTf7tuhDyyCDgkr3YIcE5+vgg4QGn2xeeTztxfkfPMakm75//b+wvbdFo7Yy3u613Ar/N1cIOsFGVvWcrZkparg1SWjlcpy+JOGOE706jMrLyIOCYito+IaaT/7a8j4n30zzEPTP7tpGZ+T3pN0nNyTxuSNgfeCIw0Imp9UY4ZVvYjzfxyG3BsTjscODyemoHlm3n5YmBGl+J6Hamr+nrg2vzYrya2I4EbSTMAXQb8dRfiekF+v+vye5fpM3sGqfLwrEJaTz+vLhzzBvm3148R8sizgQuBW/PfCYVtjs3HcAuFmSOBGaQK7G3ANwB1IN4zgBXAX0hnHg9rZ6zA04EfkiYyuQJ4Qa//R2V4lKHsLUs5W7ZydRDL0jZ8ZqUrizt0nI2+Mw3LzH56UJhRtZ+Ouez5t97vdK9jqhNj3e9Gr+OqifFlwP/kGG8APtXK9sOVGjMzMzMzMyupMgyVNDMzMzMzsxG44WZmZmZmZlZybriZmZmZmZmVnBtuZmZmZmZmJeeGm5mZmZmZWcm54WZmZmZmZlZybriZmZmZmZmVnBtuPSLpeZLWSNq4iXWnSQpJT+tGbGaSvi3pk+Pcx0xJy8aw3XxJnx/Pe5uZmZn1GzfcOkTSUklvbLQ8Iu6KiC0jYl034zKrJelQSZcU0yLi8Ij4XK9iMusVSTtJWiTpYUmrJV0k6a97HZeZmZkbbj3gnjMri6rlxarFa+UyWv6R9ELgt8Bi4PnAZOBs4DxJr+18hGZmZo254dYBkk4Fngf8vzwc8l/zUMfDJN0F/Lp2+KOkIUlfkHRFPtN7jqQJDfb/LEknSVoh6R5Jn29myKX1D0lzJf2oJu2rkr42Uv7IvWu/lfRlSQ8AZwHfBl6b8+pDeb31hitKmiXpWkmPSLpN0r45/QOSbs49E7dL+tAYjuUVkq7J+zgLeHph2UxJyyQdLele4PuSNpP0FUnL8+MrkjarWf/jku7LPd8HtfwBW9dJ+pikH9ekfT3/f0fK0y+U9GtJ9+f/+WmSti7sY2nOP9cDj0p6Wn59T85zt0jaK69+HHBpRBwbEQ9ExOqI+BpwKvDveX/DZffsnP9WSJpTeL+N8vfzthzTwuGyvLDtIZLuyvEe28nP1czM+ocbbh0QEQcDdwF/FxFbAgvzor8FXgLs02DT9wP/QDrLuxb4WoP1FuTlLwJeAbwJ+GBbgreqOAPYT9IzAXIldn/gdEbPH68Bbge2Bd4HHE6qrG4ZEVtTQ9JuwCnAx4CtgdcDS/PiVcBbgWcCHwC+LOmVzR6EpE2Bn5IqxhOAHwLvrFntuXnZDsBs4Fhgd2BX4OXAbsAnatafCEwBDgFOlAgEohIAABJ1SURBVLRjszFZz/wA2He40ZVPar2HlDdGytMCvkAqN18CTCU1wIoOBN5Cyr8vBI4EXh0RW5HK46V5vb1JebDWQmAPSc8opO0JTM+xzNVTQ+OPAt5OKu8nAw8C36zZ3+uAHYG9gE9JeknDT8XMzCxzw627jouIRyPifxssPzUiboiIR4FPAvvX9qRJmgS8GfhI3tcq4MvAAR2N3EolIu4EriFVEAHeAPwJuIPR88fyiPh6RKwdIS8WHQacHBHnR8QTEXFPRPw+x3FuRNwWyW+A84C/aeFQdgc2Ab4SEX+JiB8BV9as8wTw6Yh4PMd7EPDZiFgVEX8EPgMcXLPNJ/P6vwHOJTVqrcQiYgVwMfDunLQvcB+wjBHydEQsyXnz8ZwfTiA1moq+FhF35/yzDtgM2EnSJhGxNCJuy+tNBFbUCW8F6fdym0LaZ3I8i4HvkxqHAB8Cjo2IZRHxOKkR+S6tP0zzMxHxvxFxHXAd6QSEmZnZiHy9SHfd3cLyO0kV2ok16+yQ01dIGk7bqIl9W/85nVRZPAV4b37dTP5oNa9MBX5eb4GkNwOfBl6c3+cZpOuDmjUZuCciopB2Z806f4yIx2q2ubNm/cmF1w/mkx+Nllt5LQD+CfguqTf4VEbJ05K2JY1O+Btgq7zswZr9PpnnI2KJpI+QGlQ7S/oV8NGIWE5qKG5XJ67tSCcQHiT1VK+3T1Ie2yU/3wE4W9ITheXrgEmF1/cWnv8J2LLOe5qZma3HPW6dE02mFU0tPH8e8BdSRaLobuBxYGJEbJ0fz4yIncceqlXUD4GZkrYH/p7UcGsmf9Tmw9Hy5d2k4WXrydeV/Rj4IjApD7P8OWnoWrNWAFNUqJGT8v5I8S0nVY6L6y8vvN5G0hYjLLfy+inwMkkvJQ3BPY3R8/QXSHnkZRHxTFKDrzYPrpeHIuL0iHgdKR8F+fo14AKe6vEr2p80nPhPhbTa8no4j90NvLkQ69YR8fSIuKfZD8HMzKweN9w6ZyXwgha3eZ/SVNTPAD4L/Kj2dgF5ONF5wJckPTNfCP9CSbVDg6zP5WFhQ6RhWndExM1jzB8rge3z9Wb1nAR8QNJeeX9TJP0VsClpyNkfgbW59+1NLR7GpaRrl47Kk0a8g3TN2kjOAD4h6TmSJgKfIl0fVfQZSZtK+htSA6DedUtWMrln9UekkxBX5NumjJantwLWAA9JmkK6FrMhSTtKekM+8fAYMDx8EtKw27+WdLykCZK2kvR/SNcfH12zq09KeoaknUnXd56V078NHC9ph/x+z5E0a8wfipmZWeaGW+d8gVS5fAh4V5PbnArMJw2jeTrpIvd63k+qNN9EGrrzI+oP77H+dzrwxvx3WKv549fAjcC9kmp7eImIK8gTjwAPA78BdoiI1aQ8ujC/z3uBRa0EHxF/Bt4BHJr38R7gJ6Ns9nngKuB60rDMa3LasHvzvpaTemwOH74mzyphAWnY4amFtJHy9GeAV5Ly5rmMnn82A+aRRjPcSxr6+HGAiLiVNHHIy0kTlqwgTZazT0T8tmY/vwGWABcCX4yI83L6V0nfg/MkrQYuI00IZGZmNi5a/9IS6xVJQ8APIuJ7vY7FrKokzSR9j7bvdSw2NpKeB/weeG5EPNLreGpJmkaaBGiTiFjb22jMzGyQuMfNzMxKQdJGwEeBM8vYaDMzM+slzyppZh2Ve1BuarB4p4i4q5vxWDnlCWVWkmZo3LfH4ZiZmZWOh0qamZmZmZmVnIdKmpmVkKSTJa2SdEMhbYKk8yXdmv9uU1h2jKQlkm6RtE9vojaz/9/e/cfIcZYHHP8+mB+iLhUgk2sUp1z+cBGhEbScTFCk6tIIMKTC/EGQowJOleooSiSQIpFL/wAJCcmq1KpqRX9YJcIRkNQSpLEafoXQE/xBi+00bRKCVTeY9GorVgAlcUCgg6d/7FyyWe+eb/d2dt6d/X6k0+3OzN4+78xzo3123nlfSapLEVfcduzYkfPz8+ctf/bZZ9m+ffv5LyhAqbGVGheMFtvx48efzMzX1BTSWPTmb8nHYFza3sZxtW8r+RsRv09nmPs7MvN3qmV/Dvw4Mw9ExDLwqsy8NSIupzNNwm46k41/A/jt3ulEeg0697Zd2/N3kGHaPQ3nXkmaNUXc4zY/P8+xY8fOW76yssLi4uLkA9qEUmMrNS4YLbaI+GE90YxPb/6WfAzGpe1tHFf7tpK/mfmtagTDbnuBxerxITrz+N1aLb8rM38O/CAiTtIp4r6z0XsMOve2Xdvzd5Bh2j0N515JmjVFFG6SpE2ZqyakJjPPRMRF1fJL6MwXtm61WnaeiFgClgDm5uZYWVmpL9pCnTt3znZLkqaOhZskTb/os6xvP/jMPAgcBFhYWEivPM2OWW23JLVF0YXbQ//3FDcs3zvUa04duLamaKThzJu7Gr8nIuLi6mrbxcDZavkqcGnXdjuB0xOPbsr5PytJKpmjSkrS9DgC7K8e7wfu6Vq+LyJeFhGXAbuA7zYQnyRJqknRV9wkaVZFxJ10BiLZERGrwCeAA8DhiLgReBy4DiAzH4mIw3QmOl8DbrrQiJKSJGm6WLhJUoEy8/oBq64ZsP2ngE/VF5EkSWqSXSUlSZIkqXAWbpIkSZJUuNZ1lXRUMEmSJElt4xU3SZIkSSqchZskSZIkFa51XSWlaTVsN1+wq68kSdKs8IqbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwFm6SJEmSVDgLN0mSJEkqnIWbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4V7cdABSUyLiFPAM8EtgLTMXIuLVwD8B88Ap4H2Z+ZOmYpQkSZLAK27S1Zn5psxcqJ4vA/dn5i7g/uq5JEmS1CgLN+mF9gKHqseHgPc0GIskSZIE2FVSsy2Br0dEAv+QmQeBucw8A5CZZyLion4vjIglYAlgbm6OlZWV59adO3eOlZUVbrlire74X/C+k7TexrZqe/skSdL0sXDTLLsqM09Xxdl9EfH9zb6wKvIOAiwsLOTi4uJz61ZWVlhcXOSG5XvHHe95Tv3R4gW3qcN6G9uq7e2TJEnTx66SmlmZebr6fRa4G9gNPBERFwNUv882F6EkSZLUYeGmmRQR2yPiFeuPgbcDDwNHgP3VZvuBe5qJUJIkSXqeXSU1q+aAuyMCOv8HX8jMr0bEUeBwRNwIPA5c12CMkiRJErDFws15sDStMvMx4I19lv8IuGbyEUmSJEmDjaOrpPNgSZIkSVKN6ugquRdYrB4fAlaAW2t4H2nmzQ85cuWpA9fWFIkmyd4OkiTNnq0WbrXMg7Vu7uXUPhfWqHM1lTrPU6lxQdmxSVPo6sx8suv5em+HAxGxXD33SzNJklpiq4VbLfNgrfubz9/DXzxU7/gpo86DVeo8T6XGBWXHJrWAvR0kSWqxLVVF3fNgRcQL5sGqrrY5D5YkjV+tvR3abtDV/2F7eEzbvrPXgyRNt5ELt2ruqxdl5jNd82B9kufnwTrAFMyD5T1CmiXme2vU2tuh7QZd/b9h2P+PEXtsNMVeD5I03bZyxc15sCSpAfZ22JxBX1TccsXa0EWaJElNG7lwcx4sSZq8tvR2kCRJw6l35A9J0rjZ20GSpBlk4SZJU8TeDtNr2HtMwftMJUnPe1HTAUiSJEmSNmbhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwjio5pPVRwYaZwNVRwSSpfUYZJVKSpFF5xU2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwFm6SJEmSVDinA5A00KDhzgdNh+HUF6qLQ+9LkmadV9wkSZIkqXAWbpIkSZJUOAs3SZIkSSqc97hJGptR7kPyvjhJkqQL84qbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4SzcJEmSJKlwTgdQoGGHVHc4dU0z870dPI6SJNXLK26SJEmSVDivuEmSJm6UydolSZplXnGTJEmSpMJ5xa0Fur+5vuWKNW6o4Zts70eRJEmSmmPhJmmqOAhG/ezGKElSeWrrKhkReyLiREScjIjlut5HGjdzV9PM/JUkqZ1queIWEduATwNvA1aBoxFxJDO/V8f7qX6zcpXD3NX88r1DdzkuJd/N3/YZ59XPfnldSu5Kki6srq6Su4GTmfkYQETcBewF/PAwI/p92Njow3BBHx7MXU0z81eSpJaKzBz/H414L7AnM/+kev4B4C2ZeXPXNkvAUvX0dcCJPn9qB/Dk2AMcj1JjKzUuGC2212bma+oIpp/N5G61fKP8LfkYjEvb2ziu9hWXv5s897Zd2/N3kGHaPdHclSRdWF1X3KLPshdUiJl5EDi44R+JOJaZC+MMbFxKja3UuKDs2LpcMHdh4/ydknZuSdvbOMXtG8u5t+2m+Phuyay2W5Laoq7BSVaBS7ue7wRO1/Re0jiZu5pm5q8kSS1VV+F2FNgVEZdFxEuBfcCRmt5LGidzV9PM/JUkqaVq6SqZmWsRcTPwNWAbcHtmPjLCnyq5O0+psZUaF5QdGzC23C2+nWPQ9jZOZfvGeO5tu6k8vmMwq+2WpFaoZXASSZIkSdL41DYBtyRJkiRpPCzcJEmSJKlwRRRuEbEnIk5ExMmIWO6zPiLir6v1/xURvzehuC6NiH+NiEcj4pGI+EifbRYj4qmIeLD6+fiEYjsVEQ9V73msz/qJ77OIeF3XfngwIp6OiI/2bNPI/pqUC+VyqQblekS8OiLui4j/rn6/qus1t1XtPBER7+ha/uYqN09WOdhviPqJi4htEfEfEfEv1fPWtE2D9TtXbnTsp1lE3B4RZyPi4a5lQ+e5JKlMjRduEbEN+DTwTuBy4PqIuLxns3cCu6qfJeDvJhTeGnBLZr4euBK4qU9sAN/OzDdVP5+cUGwAV1fv2W9enonvs8w8sb4fgDcDPwXu7rNpU/urVpvM5VINyvVl4P7M3AXcXz2nWrcPeAOwB/jbqv3QybUlns+/PZNsyAY+Ajza9bxNbdPGes+VfY99C3yW83NylDyXJBWo8cIN2A2czMzHMvMXwF3A3p5t9gJ3ZMe/Aa+MiIvrDiwzz2TmA9XjZ+h86Luk7vcdk0b2WZdrgP/JzB9O8D2btplcLtIGub4XOFRtdgh4T/V4L3BXZv48M38AnAR2Vzn2G5n5neyMfHRH12saExE7gWuBf+xa3Iq2aSSDjv1Uy8xvAT/uWTxUnk8kUEnSSEoo3C4B/rfr+SrnF0eb2aZWETEP/C7w731WvzUi/jMivhIRb5hQSAl8PSKOR8RSn/VN77N9wJ0D1jWxvyah6X0+Fj25PpeZZ6BT3AEXVZsNausl1ePe5U37K+BjwK+6lrWlbdpYv3PloGPfRsPmuSSpULXM4zakfveI9M5RsJltahMRvw58EfhoZj7ds/oB4LWZeS4i3gX8M50uVHW7KjNPR8RFwH0R8f3q29bnwu7zmonss+hM/Ptu4LY+q5vaX5PQaJ6OQ2+ub3AL16C2FrcPIuIPgbOZeTwiFjfzkj7LimybNuW8c2XTARXCfJakKVPCFbdV4NKu5zuB0yNsU4uIeAmdD7Kfz8wv9a7PzKcz81z1+MvASyJiR91xZebp6vdZOveR9XZxaWyf0bnH64HMfKJ3RVP7a0Ka3OdbNiDXn1jvYlv9PlstH9TW1epx7/ImXQW8OyJO0em++gcR8Tna0TZdwIBz5aBj30bD5rkkqVAlFG5HgV0RcVl1pWYfcKRnmyPAB6PjSuCp9a4fdapGjPsM8Ghm/uWAbX5zfWS5iNhNZ5/+qOa4tkfEK9YfA28HHu7ZrJF9VrmeAd0km9hfE7SZXC7SBrl+BNhfPd4P3NO1fF9EvCwiLqNz1fS7VY49ExFXVn/zg12vaURm3paZOzNzns4x+WZmvp8WtE0b2+BcOejYt9FQed5AfJKkTWq8q2RmrkXEzcDXgG3A7Zn5SET8abX+74EvA++ic/P0T4E/nlB4VwEfAB6KiAerZX8G/FZXbO8FPhwRa8DPgH3VwAV1mgPuruqfFwNfyMyvlrDPIuLXgLcBH+pa1h1XE/trIgblcsNhbdagXD8AHI6IG4HHgesAqv/Rw8D36IxIeVNm/rJ63YfpjG73cuAr1U+J2tw2dQw6Vx6lz7GfdhFxJ7AI7IiIVeATjJbnkqQCRUs+M0uSJElSa5XQVVKSJEmStAELN0mSJEkqnIWbJEmSJBXOwk2SJEmSCmfhJkmSJEmFs3CTJEmSpMJZuEmSJElS4f4fTcC9QhRPDaEAAAAASUVORK5CYII=\n", - "text/plain": [ - "
" - ] - }, - "metadata": { - "needs_background": "light" - }, - "output_type": "display_data" - } - ], - "source": [ - "ski_data.hist(figsize=(15, 10))\n", - "plt.subplots_adjust(hspace=0.5);" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "These distributions are much better. There are clearly some skewed distributions, so keep an eye on `fastQuads`, `fastSixes`, and perhaps `trams`. These lack much variance away from 0 and may have a small number of relatively extreme values. Models failing to rate a feature as important when domain knowledge tells you it should be is an issue to look out for, as is a model being overly influenced by some extreme values. If you build a good machine learning pipeline, hopefully it will be robust to such issues, but you may also wish to consider nonlinear transformations of features." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.10 Population data" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Population and area data for the US states can be obtained from [wikipedia](https://simple.wikipedia.org/wiki/List_of_U.S._states). Listen, you should have a healthy concern about using data you \"found on the Internet\". Make sure it comes from a reputable source. This table of data is useful because it allows you to easily pull and incorporate an external data set. It also allows you to proceed with an analysis that includes state sizes and populations for your 'first cut' model. Be explicit about your source (we documented it here in this workflow) and ensure it is open to inspection. All steps are subject to review, and it may be that a client has a specific source of data they trust that you should use to rerun the analysis." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 29#\n", - "#Use pandas' `read_html` method to read the table from the URL below\n", - "states_url = 'https://simple.wikipedia.org/w/index.php?title=List_of_U.S._states&oldid=7168473'\n", - "usa_states = pd.___(___)" - ] - }, - { - "cell_type": "code", - "execution_count": 45, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "list" - ] - }, - "execution_count": 45, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "type(usa_states)" - ] - }, - { - "cell_type": "code", - "execution_count": 46, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "1" - ] - }, - "execution_count": 46, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "len(usa_states)" - ] - }, - { - "cell_type": "code", - "execution_count": 47, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
Name &postal abbs. [1]CitiesEstablished[upper-alpha 1]Population[upper-alpha 2][3]Total area[4]Land area[4]Water area[4]Numberof Reps.
Name &postal abbs. [1]Name &postal abbs. [1].1CapitalLargest[5]Established[upper-alpha 1]Population[upper-alpha 2][3]mi2km2mi2km2mi2km2Numberof Reps.
0AlabamaALMontgomeryBirminghamDec 14, 181949031855242013576750645131171177545977
1AlaskaAKJuneauAnchorageJan 3, 195973154566538417233375706411477953947432453841
2ArizonaAZPhoenixPhoenixFeb 14, 1912727871711399029523411359429420739610269
3ArkansasARLittle RockLittle RockJun 15, 183630178045317913773252035134771114329614
4CaliforniaCASacramentoLos AngelesSep 9, 18503951222316369542396715577940346679162050153
\n", - "
" + "source": [ + "missing = pd.concat([ski_data.isnull().sum(axis=1), 100 * ski_data.isnull().mean(axis=1)], axis=1)\n", + "missing.columns=['count', '%']\n", + "missing.sort_values(by='count', ascending=False).head(10)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "siL-Ka6eTNr0" + }, + "source": [ + "These seem possibly curiously quantized..." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ENwiVKCITNr0", + "outputId": "f6cb089b-314b-4381-ca0d-34107063eb70" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "array([ 0., 4., 8., 12., 16., 20.])" + ] + }, + "execution_count": 62, + "metadata": {}, + "output_type": "execute_result" + } ], - "text/plain": [ - " Name &postal abbs. [1] Cities \\\n", - " Name &postal abbs. [1] Name &postal abbs. [1].1 Capital Largest[5] \n", - "0 Alabama AL Montgomery Birmingham \n", - "1 Alaska AK Juneau Anchorage \n", - "2 Arizona AZ Phoenix Phoenix \n", - "3 Arkansas AR Little Rock Little Rock \n", - "4 California CA Sacramento Los Angeles \n", - "\n", - " Established[upper-alpha 1] Population[upper-alpha 2][3] Total area[4] \\\n", - " Established[upper-alpha 1] Population[upper-alpha 2][3] mi2 \n", - "0 Dec 14, 1819 4903185 52420 \n", - "1 Jan 3, 1959 731545 665384 \n", - "2 Feb 14, 1912 7278717 113990 \n", - "3 Jun 15, 1836 3017804 53179 \n", - "4 Sep 9, 1850 39512223 163695 \n", - "\n", - " Land area[4] Water area[4] Numberof Reps. \n", - " km2 mi2 km2 mi2 km2 Numberof Reps. \n", - "0 135767 50645 131171 1775 4597 7 \n", - "1 1723337 570641 1477953 94743 245384 1 \n", - "2 295234 113594 294207 396 1026 9 \n", - "3 137732 52035 134771 1143 2961 4 \n", - "4 423967 155779 403466 7916 20501 53 " - ] - }, - "execution_count": 47, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "usa_states = usa_states[0]\n", - "usa_states.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Note, in even the last year, the capability of `pd.read_html()` has improved. The merged cells you see in the web table are now handled much more conveniently, with 'Phoenix' now being duplicated so the subsequent columns remain aligned. But check this anyway. If you extract the established date column, you should just get dates. Recall previously you used the `.loc` accessor, because you were using labels. Now you want to refer to a column by its index position and so use `.iloc`. For a discussion on the difference use cases of `.loc` and `.iloc` refer to the [pandas documentation](https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 30#\n", - "#Use the iloc accessor to get the pandas Series for column number 4 from `usa_states`\n", - "#It should be a column of dates\n", - "established = usa_sates.___[:, 4]" - ] - }, - { - "cell_type": "code", - "execution_count": 49, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0 Dec 14, 1819\n", - "1 Jan 3, 1959\n", - "2 Feb 14, 1912\n", - "3 Jun 15, 1836\n", - "4 Sep 9, 1850\n", - "5 Aug 1, 1876\n", - "6 Jan 9, 1788\n", - "7 Dec 7, 1787\n", - "8 Mar 3, 1845\n", - "9 Jan 2, 1788\n", - "10 Aug 21, 1959\n", - "11 Jul 3, 1890\n", - "12 Dec 3, 1818\n", - "13 Dec 11, 1816\n", - "14 Dec 28, 1846\n", - "15 Jan 29, 1861\n", - "16 Jun 1, 1792\n", - "17 Apr 30, 1812\n", - "18 Mar 15, 1820\n", - "19 Apr 28, 1788\n", - "20 Feb 6, 1788\n", - "21 Jan 26, 1837\n", - "22 May 11, 1858\n", - "23 Dec 10, 1817\n", - "24 Aug 10, 1821\n", - "25 Nov 8, 1889\n", - "26 Mar 1, 1867\n", - "27 Oct 31, 1864\n", - "28 Jun 21, 1788\n", - "29 Dec 18, 1787\n", - "30 Jan 6, 1912\n", - "31 Jul 26, 1788\n", - "32 Nov 21, 1789\n", - "33 Nov 2, 1889\n", - "34 Mar 1, 1803\n", - "35 Nov 16, 1907\n", - "36 Feb 14, 1859\n", - "37 Dec 12, 1787\n", - "38 May 29, 1790\n", - "39 May 23, 1788\n", - "40 Nov 2, 1889\n", - "41 Jun 1, 1796\n", - "42 Dec 29, 1845\n", - "43 Jan 4, 1896\n", - "44 Mar 4, 1791\n", - "45 Jun 25, 1788\n", - "46 Nov 11, 1889\n", - "47 Jun 20, 1863\n", - "48 May 29, 1848\n", - "49 Jul 10, 1890\n", - "Name: (Established[upper-alpha 1], Established[upper-alpha 1]), dtype: object" - ] - }, - "execution_count": 49, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "established" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Extract the state name, population, and total area (square miles) columns." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 31#\n", - "#Now use the iloc accessor again to extract columns 0, 5, and 6 and the dataframe's `copy()` method\n", - "#Set the names of these extracted columns to 'state', 'state_population', and 'state_area_sq_miles',\n", - "#respectively.\n", - "usa_states_sub = usa_states.___[:, [___]].copy()\n", - "usa_states_sub.columns = [___]\n", - "usa_states_sub.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Do you have all the ski data states accounted for?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 32#\n", - "#Find the states in `state_summary` that are not in `usa_states_sub`\n", - "#Hint: set(list1) - set(list2) is an easy way to get items in list1 that are not in list2\n", - "missing_states = ___(state_summary.state) - ___(usa_states_sub.state)\n", - "missing_states" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "No?? " - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "If you look at the table on the web, you can perhaps start to guess what the problem is. You can confirm your suspicion by pulling out state names that _contain_ 'Massachusetts', 'Pennsylvania', or 'Virginia' from usa_states_sub:" - ] - }, - { - "cell_type": "code", - "execution_count": 52, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "20 Massachusetts[upper-alpha 3]\n", - "37 Pennsylvania[upper-alpha 3]\n", - "38 Rhode Island[upper-alpha 4]\n", - "45 Virginia[upper-alpha 3]\n", - "47 West Virginia\n", - "Name: state, dtype: object" - ] - }, - "execution_count": 52, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "usa_states_sub.state[usa_states_sub.state.str.contains('Massachusetts|Pennsylvania|Rhode Island|Virginia')]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Delete square brackets and their contents and try again:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 33#\n", - "#Use pandas' Series' `replace()` method to replace anything within square brackets (including the brackets)\n", - "#with the empty string. Do this inplace, so you need to specify the arguments:\n", - "#to_replace='\\[.*\\]' #literal square bracket followed by anything or nothing followed by literal closing bracket\n", - "#value='' #empty string as replacement\n", - "#regex=True #we used a regex in our `to_replace` argument\n", - "#inplace=True #Do this \"in place\"\n", - "usa_states_sub.state.___(to_replace=___, value=__, regex=___, inplace=___)\n", - "usa_states_sub.state[usa_states_sub.state.str.contains('Massachusetts|Pennsylvania|Rhode Island|Virginia')]" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 34#\n", - "#And now verify none of our states are missing by checking that there are no states in\n", - "#state_summary that are not in usa_states_sub (as earlier using `set()`)\n", - "missing_states = ___(state_summary.state) - ___(usa_states_sub.state)\n", - "missing_states" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Better! You have an empty set for missing states now. You can confidently add the population and state area columns to the ski resort data." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 35#\n", - "#Use 'state_summary's `merge()` method to combine our new data in 'usa_states_sub'\n", - "#specify the arguments how='left' and on='state'\n", - "state_summary = state_summary.___(usa_states_sub, ___=___, ___=___)\n", - "state_summary.head()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Having created this data frame of summary statistics for various states, it would seem obvious to join this with the ski resort data to augment it with this additional data. You will do this, but not now. In the next notebook you will be exploring the data, including the relationships between the states. For that you want a separate row for each state, as you have here, and joining the data this soon means you'd need to separate and eliminate redundances in the state data when you wanted it." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.11 Target Feature" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Finally, what will your target be when modelling ticket price? What relationship is there between weekday and weekend prices?" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 36#\n", - "#Use ski_data's `plot()` method to create a scatterplot (kind='scatter') with 'AdultWeekday' on the x-axis and\n", - "#'AdultWeekend' on the y-axis\n", - "ski_data.___(x=___, y=___, kind=___);" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "A couple of observations can be made. Firstly, there is a clear line where weekend and weekday prices are equal. Weekend prices being higher than weekday prices seem restricted to sub $100 resorts. Recall from the boxplot earlier that the distribution for weekday and weekend prices in Montana seemed equal. Is this confirmed in the actual data for each resort? Big Mountain resort is in Montana, so the relationship between these quantities in this state are particularly relevant." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "#Code task 37#\n", - "#Use the loc accessor on ski_data to print the 'AdultWeekend' and 'AdultWeekday' columns for Montana only\n", - "ski_data.___[ski_data.state == ___, [___, ___]]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Is there any reason to prefer weekend or weekday prices? Which is missing the least?" - ] - }, - { - "cell_type": "code", - "execution_count": 58, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "AdultWeekend 4\n", - "AdultWeekday 7\n", - "dtype: int64" - ] - }, - "execution_count": 58, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data[['AdultWeekend', 'AdultWeekday']].isnull().sum()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Weekend prices have the least missing values of the two, so drop the weekday prices and then keep just the rows that have weekend price." - ] - }, - { - "cell_type": "code", - "execution_count": 59, - "metadata": {}, - "outputs": [], - "source": [ - "ski_data.drop(columns='AdultWeekday', inplace=True)\n", - "ski_data.dropna(subset=['AdultWeekend'], inplace=True)" - ] - }, - { - "cell_type": "code", - "execution_count": 60, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "(277, 25)" - ] - }, - "execution_count": 60, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "ski_data.shape" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Perform a final quick check on the data." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### 2.11.1 Number Of Missing Values By Row - Resort" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Having dropped rows missing the desired target ticket price, what degree of missingness do you have for the remaining rows?" - ] - }, - { - "cell_type": "code", - "execution_count": 61, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
count%
329520.0
62520.0
141520.0
86520.0
74520.0
146520.0
184416.0
108416.0
198416.0
39416.0
\n", - "
" + "source": [ + "missing['%'].unique()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "N7_L9U2FTNr0" + }, + "source": [ + "Yes, the percentage of missing values per row appear in multiples of 4." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qU_2tu7oTNr0", + "outputId": "1f552e29-bc0f-465e-8333-b5411fc297e8" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "0.0 107\n", + "4.0 94\n", + "8.0 45\n", + "12.0 15\n", + "16.0 10\n", + "20.0 6\n", + "Name: %, dtype: int64" + ] + }, + "execution_count": 63, + "metadata": {}, + "output_type": "execute_result" + } ], - "text/plain": [ - " count %\n", - "329 5 20.0\n", - "62 5 20.0\n", - "141 5 20.0\n", - "86 5 20.0\n", - "74 5 20.0\n", - "146 5 20.0\n", - "184 4 16.0\n", - "108 4 16.0\n", - "198 4 16.0\n", - "39 4 16.0" - ] - }, - "execution_count": 61, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "missing = pd.concat([ski_data.isnull().sum(axis=1), 100 * ski_data.isnull().mean(axis=1)], axis=1)\n", - "missing.columns=['count', '%']\n", - "missing.sort_values(by='count', ascending=False).head(10)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "These seem possibly curiously quantized..." - ] - }, - { - "cell_type": "code", - "execution_count": 62, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "array([ 0., 4., 8., 12., 16., 20.])" - ] - }, - "execution_count": 62, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "missing['%'].unique()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Yes, the percentage of missing values per row appear in multiples of 4." - ] - }, - { - "cell_type": "code", - "execution_count": 63, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "0.0 107\n", - "4.0 94\n", - "8.0 45\n", - "12.0 15\n", - "16.0 10\n", - "20.0 6\n", - "Name: %, dtype: int64" - ] - }, - "execution_count": 63, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "missing['%'].value_counts()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "This is almost as if values have been removed artificially... Nevertheless, what you don't know is how useful the missing features are in predicting ticket price. You shouldn't just drop rows that are missing several useless features." - ] - }, - { - "cell_type": "code", - "execution_count": 64, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\n", - "Int64Index: 277 entries, 0 to 329\n", - "Data columns (total 25 columns):\n", - " # Column Non-Null Count Dtype \n", - "--- ------ -------------- ----- \n", - " 0 Name 277 non-null object \n", - " 1 Region 277 non-null object \n", - " 2 state 277 non-null object \n", - " 3 summit_elev 277 non-null int64 \n", - " 4 vertical_drop 277 non-null int64 \n", - " 5 base_elev 277 non-null int64 \n", - " 6 trams 277 non-null int64 \n", - " 7 fastSixes 277 non-null int64 \n", - " 8 fastQuads 277 non-null int64 \n", - " 9 quad 277 non-null int64 \n", - " 10 triple 277 non-null int64 \n", - " 11 double 277 non-null int64 \n", - " 12 surface 277 non-null int64 \n", - " 13 total_chairs 277 non-null int64 \n", - " 14 Runs 274 non-null float64\n", - " 15 TerrainParks 233 non-null float64\n", - " 16 LongestRun_mi 272 non-null float64\n", - " 17 SkiableTerrain_ac 275 non-null float64\n", - " 18 Snow Making_ac 240 non-null float64\n", - " 19 daysOpenLastYear 233 non-null float64\n", - " 20 yearsOpen 277 non-null float64\n", - " 21 averageSnowfall 268 non-null float64\n", - " 22 AdultWeekend 277 non-null float64\n", - " 23 projectedDaysOpen 236 non-null float64\n", - " 24 NightSkiing_ac 163 non-null float64\n", - "dtypes: float64(11), int64(11), object(3)\n", - "memory usage: 56.3+ KB\n" - ] + "source": [ + "missing['%'].value_counts()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Jr_MGWs1TNr1" + }, + "source": [ + "This is almost as if values have been removed artificially... Nevertheless, what you don't know is how useful the missing features are in predicting ticket price. You shouldn't just drop rows that are missing several useless features." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "F1n4Bp1LTNr1", + "outputId": "730079a1-0eec-4801-c663-b8d373bd6ba4" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Int64Index: 277 entries, 0 to 329\n", + "Data columns (total 25 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Name 277 non-null object \n", + " 1 Region 277 non-null object \n", + " 2 state 277 non-null object \n", + " 3 summit_elev 277 non-null int64 \n", + " 4 vertical_drop 277 non-null int64 \n", + " 5 base_elev 277 non-null int64 \n", + " 6 trams 277 non-null int64 \n", + " 7 fastSixes 277 non-null int64 \n", + " 8 fastQuads 277 non-null int64 \n", + " 9 quad 277 non-null int64 \n", + " 10 triple 277 non-null int64 \n", + " 11 double 277 non-null int64 \n", + " 12 surface 277 non-null int64 \n", + " 13 total_chairs 277 non-null int64 \n", + " 14 Runs 274 non-null float64\n", + " 15 TerrainParks 233 non-null float64\n", + " 16 LongestRun_mi 272 non-null float64\n", + " 17 SkiableTerrain_ac 275 non-null float64\n", + " 18 Snow Making_ac 240 non-null float64\n", + " 19 daysOpenLastYear 233 non-null float64\n", + " 20 yearsOpen 277 non-null float64\n", + " 21 averageSnowfall 268 non-null float64\n", + " 22 AdultWeekend 277 non-null float64\n", + " 23 projectedDaysOpen 236 non-null float64\n", + " 24 NightSkiing_ac 163 non-null float64\n", + "dtypes: float64(11), int64(11), object(3)\n", + "memory usage: 56.3+ KB\n" + ] + } + ], + "source": [ + "ski_data.info()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "0xg7fo_NTNr1" + }, + "source": [ + "There are still some missing values, and it's good to be aware of this, but leave them as is for now." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "NgOzPfBPTNr1" + }, + "source": [ + "## 2.12 Save data" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "a97CMhmgTNr1", + "outputId": "ade467d1-6957-4981-907f-bf2b56f530a2" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(277, 25)" + ] + }, + "execution_count": 65, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "ski_data.shape" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hQBc-p6NTNr1" + }, + "source": [ + "Save this to your data directory, separately. Note that you were provided with the data in `raw_data` and you should saving derived data in a separate location. This guards against overwriting our original data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "yioTvOBHTNr1" + }, + "outputs": [], + "source": [ + "# save the data to a new csv file\n", + "datapath = '../data'\n", + "save_file(ski_data, 'ski_data_cleaned.csv', datapath)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "OT5Y6rsnTNr2" + }, + "outputs": [], + "source": [ + "# save the state_summary separately.\n", + "datapath = '../data'\n", + "save_file(state_summary, 'state_summary.csv', datapath)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "L9N77RAWTNr2" + }, + "source": [ + "## 2.13 Summary" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "65C5KgRKTNr2" + }, + "source": [ + "**Q: 3** Write a summary statement that highlights the key processes and findings from this notebook. This should include information such as the original number of rows in the data, whether our own resort was actually present etc. What columns, if any, have been removed? Any rows? Summarise the reasons why. Were any other issues found? What remedial actions did you take? State where you are in the project. Can you confirm what the target feature is for your desire to predict ticket price? How many rows were left in the data? Hint: this is a great opportunity to reread your notebook, check all cells have been executed in order and from a \"blank slate\" (restarting the kernel will do this), and that your workflow makes sense and follows a logical pattern. As you do this you can pull out salient information for inclusion in this summary. Thus, this section will provide an important overview of \"what\" and \"why\" without having to dive into the \"how\" or any unproductive or inconclusive steps along the way." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UgTOE1phTNr2" + }, + "source": [ + "**A: 3** There were multiple different processes used in gathering information for the result. Many pandas and matplotlib methods were used as well as a few seaborn methods. There were, originally, 58 rows of data. By the end, there were 67 rows of data. I do not see Big Mountain Resort present on any of the output data. We did drop weekday prices as well as the 'fastEight' column. Some rows were also dropped because they were missing the desired target ticket price. The percentage of missing values per row appear in multiples of 4, almost as if they were removed artificially. Values were sorted and columns were not dropped for the sole purpose of missing data. I am in the Data Wrangling section of the project currently in this summary, I am started also in the Exploratory Data Analysis section. The target feature for the desire to predict ticket price, in my opinion, is whether it is a weekday or weekend and how much skiable terrain is available.\n", + "\n", + "Any constructive criticism is greatly appreciated as I am still very new to Data Science, but loving it so far! Thank you so much!" + ] } - ], - "source": [ - "ski_data.info()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "There are still some missing values, and it's good to be aware of this, but leave them as is for now." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.12 Save data" - ] - }, - { - "cell_type": "code", - "execution_count": 65, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "(277, 25)" - ] - }, - "execution_count": 65, - "metadata": {}, - "output_type": "execute_result" + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.4" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": true, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true + }, + "varInspector": { + "cols": { + "lenName": 16, + "lenType": 16, + "lenVar": 40 + }, + "kernels_config": { + "python": { + "delete_cmd_postfix": "", + "delete_cmd_prefix": "del ", + "library": "var_list.py", + "varRefreshCmd": "print(var_dic_list())" + }, + "r": { + "delete_cmd_postfix": ") ", + "delete_cmd_prefix": "rm(", + "library": "var_list.r", + "varRefreshCmd": "cat(var_dic_list()) " + } + }, + "types_to_exclude": [ + "module", + "function", + "builtin_function_or_method", + "instance", + "_Feature" + ], + "window_display": false + }, + "colab": { + "provenance": [] } - ], - "source": [ - "ski_data.shape" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Save this to your data directory, separately. Note that you were provided with the data in `raw_data` and you should saving derived data in a separate location. This guards against overwriting our original data." - ] }, - { - "cell_type": "code", - "execution_count": 66, - "metadata": {}, - "outputs": [], - "source": [ - "# save the data to a new csv file\n", - "datapath = '../data'\n", - "save_file(ski_data, 'ski_data_cleaned.csv', datapath)" - ] - }, - { - "cell_type": "code", - "execution_count": 67, - "metadata": {}, - "outputs": [], - "source": [ - "# save the state_summary separately.\n", - "datapath = '../data'\n", - "save_file(state_summary, 'state_summary.csv', datapath)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2.13 Summary" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**Q: 3** Write a summary statement that highlights the key processes and findings from this notebook. This should include information such as the original number of rows in the data, whether our own resort was actually present etc. What columns, if any, have been removed? Any rows? Summarise the reasons why. Were any other issues found? What remedial actions did you take? State where you are in the project. Can you confirm what the target feature is for your desire to predict ticket price? How many rows were left in the data? Hint: this is a great opportunity to reread your notebook, check all cells have been executed in order and from a \"blank slate\" (restarting the kernel will do this), and that your workflow makes sense and follows a logical pattern. As you do this you can pull out salient information for inclusion in this summary. Thus, this section will provide an important overview of \"what\" and \"why\" without having to dive into the \"how\" or any unproductive or inconclusive steps along the way." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "**A: 3** Your answer here" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.7.4" - }, - "toc": { - "base_numbering": 1, - "nav_menu": {}, - "number_sections": true, - "sideBar": true, - "skip_h1_title": false, - "title_cell": "Table of Contents", - "title_sidebar": "Contents", - "toc_cell": false, - "toc_position": {}, - "toc_section_display": true, - "toc_window_display": true - }, - "varInspector": { - "cols": { - "lenName": 16, - "lenType": 16, - "lenVar": 40 - }, - "kernels_config": { - "python": { - "delete_cmd_postfix": "", - "delete_cmd_prefix": "del ", - "library": "var_list.py", - "varRefreshCmd": "print(var_dic_list())" - }, - "r": { - "delete_cmd_postfix": ") ", - "delete_cmd_prefix": "rm(", - "library": "var_list.r", - "varRefreshCmd": "cat(var_dic_list()) " - } - }, - "types_to_exclude": [ - "module", - "function", - "builtin_function_or_method", - "instance", - "_Feature" - ], - "window_display": false - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file diff --git a/Notebooks/03_exploratory_data_analysis.ipynb b/Notebooks/03_exploratory_data_analysis.ipynb index c1746d2e4..7de859e3a 100644 --- a/Notebooks/03_exploratory_data_analysis.ipynb +++ b/Notebooks/03_exploratory_data_analysis.ipynb @@ -994,7 +994,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1018,7 +1018,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1158,11 +1158,11 @@ "source": [ "#Code task 1#\n", "#Create a new dataframe, `state_summary_scale` from `state_summary` whilst setting the index to 'state'\n", - "state_summary_scale = state_summary.set_index(___)\n", + "state_summary_scale = state_summary.set_index('state')\n", "#Save the state labels (using the index attribute of `state_summary_scale`) into the variable 'state_summary_index'\n", - "state_summary_index = state_summary_scale.___\n", + "state_summary_index = state_summary_scale.index\n", "#Save the column names (using the `columns` attribute) of `state_summary_scale` into the variable 'state_summary_columns'\n", - "state_summary_columns = state_summary_scale.___\n", + "state_summary_columns = state_summary_scale.tolist()\n", "state_summary_scale.head()" ] }, @@ -1199,7 +1199,7 @@ "source": [ "#Code task 2#\n", "#Create a new dataframe from `state_summary_scale` using the column names we saved in `state_summary_columns`\n", - "state_summary_scaled_df = pd.DataFrame(___, columns=___)\n", + "state_summary_scaled_df = pd.DataFrame(states, columns='state_summary_columns')\n", "state_summary_scaled_df.head()" ] }, @@ -1232,7 +1232,7 @@ "source": [ "#Code task 3#\n", "#Call `state_summary_scaled_df`'s `mean()` method\n", - "state_summary_scaled_df.___" + "state_summary_scaled_df.mean()" ] }, { @@ -1257,7 +1257,7 @@ "source": [ "#Code task 4#\n", "#Call `state_summary_scaled_df`'s `std()` method\n", - "state_summary_scaled_df.___" + "state_summary_scaled_df.std()" ] }, { @@ -1277,7 +1277,7 @@ "source": [ "#Code task 5#\n", "#Repeat the previous call to `std()` but pass in ddof=0 \n", - "state_summary_scaled_df.___(___)" + "state_summary_scaled_df.std(ddof=0)" ] }, { @@ -1330,10 +1330,10 @@ "#title to 'Cumulative variance ratio explained by PCA components for state/resort summary statistics'\n", "#Hint: remember the handy ';' at the end of the last plot call to suppress that untidy output\n", "plt.subplots(figsize=(10, 6))\n", - "plt.plot(state_pca.explained_variance_ratio_.___)\n", - "plt.xlabel(___)\n", - "plt.ylabel(___)\n", - "plt.title(___);" + "plt.plot(state_pca.explained_variance_ratio_.cumsum())\n", + "plt.xlabel('Component #')\n", + "plt.ylabel('Cumulative ratio variance')\n", + "plt.title('Cumulative variance ratio explained by PCA components for state/resort summary statistics');" ] }, { @@ -1365,7 +1365,7 @@ "source": [ "#Code task 7#\n", "#Call `state_pca`'s `transform()` method, passing in `state_summary_scale` as its argument\n", - "state_pca_x = state_pca.___(___)" + "state_pca_x = state_pca.transform(`state_summary_scale`)" ] }, { @@ -1411,7 +1411,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1458,7 +1458,7 @@ "source": [ "#Code task 8#\n", "#Calculate the average 'AdultWeekend' ticket price by state\n", - "state_avg_price = ski_data.groupby(___)[___].___\n", + "state_avg_price = ski_data.groupby('AdultWeekend')['price'].mean()\n", "state_avg_price.head()" ] }, @@ -1469,7 +1469,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1512,7 +1512,7 @@ "#Remember the first component was given by state_pca_x[:, 0],\n", "#and the second by state_pca_x[:, 1]\n", "#Call these 'PC1' and 'PC2', respectively and set the dataframe index to `state_summary_index`\n", - "pca_df = pd.DataFrame({'PC1': ___, 'PC2': ___}, index=__)\n", + "pca_df = pd.DataFrame({'PC1': [:, 0], 'PC2': [:, 1]}, index=`state_summary_index`)\n", "pca_df.head()" ] }, @@ -1644,7 +1644,7 @@ "#Code task 10#\n", "#Use pd.concat to concatenate `pca_df` and `state_avg_price` along axis 1\n", "# remember, pd.concat will align on index\n", - "pca_df = ___([___, ___], axis=___)\n", + "pca_df = pd.concat([`pca_df`, `state_avg_price`], axis=1)\n", "pca_df.head()" ] }, @@ -1901,7 +1901,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -1966,8 +1966,8 @@ "plt.subplots(figsize=(12, 10))\n", "# Note the argument below to make sure we get the colours in the ascending\n", "# order we intuitively expect!\n", - "sns.___(x=___, y=___, size=___, hue=___, \n", - " hue_order=___, data=pca_df)\n", + "sns.scatterplot(x=pca_df.PC1, y=pca_df.PC2, size='AdultWeekend', hue='Quartile', \n", + " hue_order=pca_df.Quartile.cat.categories, data=pca_df)\n", "#and we can still annotate with the state labels\n", "for s, x, y in zip(state, x, y):\n", " plt.annotate(s, (x, y)) \n", @@ -3301,7 +3301,7 @@ "#Show a seaborn heatmap of correlations in ski_data\n", "#Hint: call pandas' `corr()` method on `ski_data` and pass that into `sns.heatmap`\n", "plt.subplots(figsize=(12,10))\n", - "sns.___(ski_data.___);" + "sns.heatmap(ski_data.corr());" ] }, { @@ -3362,7 +3362,7 @@ "#Code task 13#\n", "#Use a list comprehension to build a list of features from the columns of `ski_data` that\n", "#are _not_ any of 'Name', 'Region', 'state', or 'AdultWeekend'\n", - "features = [___ for ___ in ski_data.columns if ___ not in [___, ___, ___, ___]]" + "features = [col for col in ski_data.columns if col not in ['Name', 'Region', 'state', 'AdultWeekend']]" ] }, { @@ -3372,7 +3372,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -3420,7 +3420,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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", "text/plain": [ "
" ] @@ -3463,7 +3463,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "**A: 1** Your answer here" + "**A: 1** I found there were many numerical data features, such as: Total state area, Total state population, Resorts per state, Total skiable area, Total night skiing area, Total days open, Resort density, Average ticket price by state and an Average ticket price scatterplot. I did not find as many categorical data features, however, I was able to find a couple; Top states by order of each of the summary statistics, Top states by resort density. I could not find a specific pattern suggesting a relationship between state and ticket price. I was lead to the conclusion that seaborn is the most comprehensive feature to use regarding subsequent modeling. I've found to always remain wary of the following aspects when performing feature selection: Multicollinearity, Irrelevant features, Overfitting, Feature scaling, Data leakage, Missing values, Feature interaction, Domain knowledge, and Dimensionality." ] }, { diff --git a/Notebooks/Copy of 05_modeling.ipynb b/Notebooks/Copy of 05_modeling.ipynb new file mode 100644 index 000000000..ca8b7bb2b --- /dev/null +++ b/Notebooks/Copy of 05_modeling.ipynb @@ -0,0 +1 @@ +{"cells":[{"cell_type":"markdown","metadata":{"id":"r6ISxyT2XvBL"},"source":["# 5 Modeling"]},{"cell_type":"markdown","metadata":{"id":"2XKgHpiPXvBO"},"source":["## 5.1 Contents\n","* [5 Modeling](#5_Modeling)\n"," * [5.1 Contents](#5.1_Contents)\n"," * [5.2 Introduction](#5.2_Introduction)\n"," * [5.3 Imports](#5.3_Imports)\n"," * [5.4 Load Model](#5.4_Load_Model)\n"," * [5.5 Load Data](#5.5_Load_Data)\n"," * [5.6 Refit Model On All Available Data (excluding Big Mountain)](#5.6_Refit_Model_On_All_Available_Data_(excluding_Big_Mountain))\n"," * [5.7 Calculate Expected Big Mountain Ticket Price From The Model](#5.7_Calculate_Expected_Big_Mountain_Ticket_Price_From_The_Model)\n"," * [5.8 Big Mountain Resort In Market Context](#5.8_Big_Mountain_Resort_In_Market_Context)\n"," * [5.8.1 Ticket price](#5.8.1_Ticket_price)\n"," * [5.8.2 Vertical drop](#5.8.2_Vertical_drop)\n"," * [5.8.3 Snow making area](#5.8.3_Snow_making_area)\n"," * [5.8.4 Total number of chairs](#5.8.4_Total_number_of_chairs)\n"," * [5.8.5 Fast quads](#5.8.5_Fast_quads)\n"," * [5.8.6 Runs](#5.8.6_Runs)\n"," * [5.8.7 Longest run](#5.8.7_Longest_run)\n"," * [5.8.8 Trams](#5.8.8_Trams)\n"," * [5.8.9 Skiable terrain area](#5.8.9_Skiable_terrain_area)\n"," * [5.9 Modeling scenarios](#5.9_Modeling_scenarios)\n"," * [5.9.1 Scenario 1](#5.9.1_Scenario_1)\n"," * [5.9.2 Scenario 2](#5.9.2_Scenario_2)\n"," * [5.9.3 Scenario 3](#5.9.3_Scenario_3)\n"," * [5.9.4 Scenario 4](#5.9.4_Scenario_4)\n"," * [5.10 Summary](#5.10_Summary)\n"," * [5.11 Further work](#5.11_Further_work)\n"]},{"cell_type":"markdown","metadata":{"id":"5orEnEkCXvBP"},"source":["## 5.2 Introduction"]},{"cell_type":"markdown","metadata":{"id":"xdD-fo8tXvBP"},"source":["In this notebook, we now take our model for ski resort ticket price and leverage it to gain some insights into what price Big Mountain's facilities might actually support as well as explore the sensitivity of changes to various resort parameters. Note that this relies on the implicit assumption that all other resorts are largely setting prices based on how much people value certain facilities. Essentially this assumes prices are set by a free market.\n","\n","We can now use our model to gain insight into what Big Mountain's ideal ticket price could/should be, and how that might change under various scenarios."]},{"cell_type":"markdown","metadata":{"id":"W84v0ZrjXvBQ"},"source":["## 5.3 Imports"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Qd1mQvG9XvBQ"},"outputs":[],"source":["import pandas as pd\n","import numpy as np\n","import os\n","import pickle\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","from sklearn import __version__ as sklearn_version\n","from sklearn.model_selection import cross_validate"]},{"cell_type":"markdown","metadata":{"id":"cchLgSt0XvBR"},"source":["## 5.4 Load Model"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1b2OAqacXvBS","executionInfo":{"status":"ok","timestamp":1721138649551,"user_tz":240,"elapsed":194,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"63ebb8a3-b585-4ad8-c206-5e6be79c0d4e"},"outputs":[{"output_type":"stream","name":"stdout","text":["Expected model not found\n"]}],"source":["# This isn't exactly production-grade, but a quick check for development\n","# These checks can save some head-scratching in development when moving from\n","# one python environment to another, for example\n","expected_model_version = '1.0'\n","model_path = '../models/ski_resort_pricing_model.pkl'\n","if os.path.exists(model_path):\n"," with open(model_path, 'rb') as f:\n"," model = pickle.load(f)\n"," if model.version != expected_model_version:\n"," print(\"Expected model version doesn't match version loaded\")\n"," if model.sklearn_version != sklearn_version:\n"," print(\"Warning: model created under different sklearn version\")\n","else:\n"," print(\"Expected model not found\")"]},{"cell_type":"markdown","metadata":{"id":"y9dyg7T1XvBS"},"source":["## 5.5 Load Data"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"NDgmXQ2DXvBS"},"outputs":[],"source":["ski_data = pd.read_csv('https://raw.githubusercontent.com/JLindsey96/DataScienceGuidedCapstone/master/raw_data/ski_resort_data.csv')"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"qpOjDCqKXvBT"},"outputs":[],"source":["big_mountain = ski_data[ski_data.Name == 'Big Mountain Resort']"]},{"cell_type":"code","execution_count":null,"metadata":{"scrolled":true,"colab":{"base_uri":"https://localhost:8080/","height":896},"id":"Pq0I8LmeXvBT","executionInfo":{"status":"ok","timestamp":1721138655685,"user_tz":240,"elapsed":179,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"4e514691-3ea3-44b9-f943-b160a39c9ed1"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" 151\n","Name Big Mountain Resort\n","Region Montana\n","state Montana\n","summit_elev 6817\n","vertical_drop 2353\n","base_elev 4464\n","trams 0\n","fastEight 0.0\n","fastSixes 0\n","fastQuads 3\n","quad 2\n","triple 6\n","double 0\n","surface 3\n","total_chairs 14\n","Runs 105.0\n","TerrainParks 4.0\n","LongestRun_mi 3.3\n","SkiableTerrain_ac 3000.0\n","Snow Making_ac 600.0\n","daysOpenLastYear 123.0\n","yearsOpen 72.0\n","averageSnowfall 333.0\n","AdultWeekday 81.0\n","AdultWeekend 81.0\n","projectedDaysOpen 123.0\n","NightSkiing_ac 600.0"],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
151
NameBig Mountain Resort
RegionMontana
stateMontana
summit_elev6817
vertical_drop2353
base_elev4464
trams0
fastEight0.0
fastSixes0
fastQuads3
quad2
triple6
double0
surface3
total_chairs14
Runs105.0
TerrainParks4.0
LongestRun_mi3.3
SkiableTerrain_ac3000.0
Snow Making_ac600.0
daysOpenLastYear123.0
yearsOpen72.0
averageSnowfall333.0
AdultWeekday81.0
AdultWeekend81.0
projectedDaysOpen123.0
NightSkiing_ac600.0
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n"," \n","\n","\n","\n"," \n","
\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"big_mountain\",\n \"rows\": 27,\n \"fields\": [\n {\n \"column\": 151,\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 19,\n \"samples\": [\n \"Big Mountain Resort\",\n 0,\n 4.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":5}],"source":["big_mountain.T"]},{"cell_type":"markdown","metadata":{"id":"QVlVPMSUXvBU"},"source":["## 5.6 Refit Model On All Available Data (excluding Big Mountain)"]},{"cell_type":"markdown","metadata":{"id":"0BjP7G3lXvBU"},"source":["This next step requires some careful thought. We want to refit the model using all available data. But should we include Big Mountain data? On the one hand, we are _not_ trying to estimate model performance on a previously unseen data sample, so theoretically including Big Mountain data should be fine. One might first think that including Big Mountain in the model training would, if anything, improve model performance in predicting Big Mountain's ticket price. But here's where our business context comes in. The motivation for this entire project is based on the sense that Big Mountain needs to adjust its pricing. One way to phrase this problem: we want to train a model to predict Big Mountain's ticket price based on data from _all the other_ resorts! We don't want Big Mountain's current price to bias this. We want to calculate a price based only on its competitors."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"du2gn7zzXvBU"},"outputs":[],"source":["# Assuming 'model' is an object with an attribute 'X_columns',\n","# you need to define 'model' before using it.\n","# For example, if 'model' is a scikit-learn model:\n","\n","from sklearn.linear_model import LinearRegression\n","\n","# Initialize the model\n","model = LinearRegression()\n","\n","# Define the columns you want to use as features\n","# Replace with the actual names of columns you want to use\n","model.X_columns = [\"summit_elev\", \"vertical_drop\", \"trams\", \"fastEight\"]\n","\n","# Now you can use the 'model' object\n","X = ski_data.loc[ski_data.Name != \"Big Mountain Resort\", model.X_columns]\n","y = ski_data.loc[ski_data.Name != \"Big Mountain Resort\", 'AdultWeekend']"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Q7p1j3bNXvBU","executionInfo":{"status":"ok","timestamp":1721138666337,"user_tz":240,"elapsed":184,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"8aa72191-8241-482b-f480-1dbfbae831b7"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["(329, 329)"]},"metadata":{},"execution_count":7}],"source":["len(X), len(y)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":75},"id":"e_w7Ac74XvBU","executionInfo":{"status":"ok","timestamp":1721138677770,"user_tz":240,"elapsed":147,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"73fb0094-61bc-4bdd-d1e5-cb9216c35dfe"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["LinearRegression()"],"text/html":["
LinearRegression()
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
"]},"metadata":{},"execution_count":8}],"source":["# Assuming 'ski_data' is a pandas DataFrame\n","import pandas as pd\n","from sklearn.linear_model import LinearRegression\n","\n","# Initialize the model\n","model = LinearRegression()\n","\n","# Define the columns you want to use as features\n","model.X_columns = [\"summit_elev\", \"vertical_drop\", \"trams\", \"fastEight\"]\n","\n","# Handle missing values (NaN) in 'ski_data'\n","# Option 1: Drop rows with missing values in BOTH X and y\n","ski_data_cleaned = ski_data.dropna(subset=model.X_columns + ['AdultWeekend'])\n","\n","# Option 2: Fill missing values with a specific value (e.g., 0) in BOTH X and y\n","# ski_data_cleaned = ski_data.fillna(0)\n","\n","# Now you can use the 'model' object with the cleaned data\n","X = ski_data_cleaned.loc[ski_data_cleaned.Name != \"Big Mountain Resort\", model.X_columns]\n","y = ski_data_cleaned.loc[ski_data_cleaned.Name != \"Big Mountain Resort\", 'AdultWeekend']\n","\n","# Fit the model\n","model.fit(X, y)"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"mqKnM6zfXvBU"},"outputs":[],"source":["cv_results = cross_validate(model, X, y, scoring='neg_mean_absolute_error', cv=5, n_jobs=-1)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Fua9q6edXvBV","executionInfo":{"status":"ok","timestamp":1721138688955,"user_tz":240,"elapsed":167,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"11155a85-6d01-4418-e6e1-ae3c41e5c2a6"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([-15.91489073, -11.35405252, -13.66349 , -13.94094958,\n"," -16.69944277])"]},"metadata":{},"execution_count":10}],"source":["cv_results['test_score']"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"OZL0Svh_XvBV","executionInfo":{"status":"ok","timestamp":1721138690822,"user_tz":240,"elapsed":151,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"5f699c63-a6d5-4b60-eb2a-8c4cfda59ca8"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["(14.314565119673142, 1.8749296358815086)"]},"metadata":{},"execution_count":11}],"source":["mae_mean, mae_std = np.mean(-1 * cv_results['test_score']), np.std(-1 * cv_results['test_score'])\n","mae_mean, mae_std"]},{"cell_type":"markdown","metadata":{"id":"RMxod_L1XvBV"},"source":["These numbers will inevitably be different to those in the previous step that used a different training data set. They should, however, be consistent. It's important to appreciate that estimates of model performance are subject to the noise and uncertainty of data!"]},{"cell_type":"markdown","metadata":{"id":"w1uzOY9fXvBV"},"source":["## 5.7 Calculate Expected Big Mountain Ticket Price From The Model"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Lma2-VIEXvBV"},"outputs":[],"source":["X_bm = ski_data.loc[ski_data.Name == \"Big Mountain Resort\", model.X_columns]\n","y_bm = ski_data.loc[ski_data.Name == \"Big Mountain Resort\", 'AdultWeekend']"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"iAu8BCwoXvBV"},"outputs":[],"source":["bm_pred = model.predict(X_bm).item()"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"RCH-2580XvBW"},"outputs":[],"source":["y_bm = y_bm.values.item()"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Db4ik3C2XvBW","executionInfo":{"status":"ok","timestamp":1721138729667,"user_tz":240,"elapsed":148,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"21a4518f-f2bb-4ed5-bacb-5656db2142ca"},"outputs":[{"output_type":"stream","name":"stdout","text":["Big Mountain Resort modelled price is $82.53, actual price is $81.00.\n","Even with the expected mean absolute error of $14.31, this suggests there is room for an increase.\n"]}],"source":["print(f'Big Mountain Resort modelled price is ${bm_pred:.2f}, actual price is ${y_bm:.2f}.')\n","print(f'Even with the expected mean absolute error of ${mae_mean:.2f}, this suggests there is room for an increase.')"]},{"cell_type":"markdown","metadata":{"id":"tn9lbbzkXvBW"},"source":["This result should be looked at optimistically and doubtfully! The validity of our model lies in the assumption that other resorts accurately set their prices according to what the market (the ticket-buying public) supports. The fact that our resort seems to be charging that much less that what's predicted suggests our resort might be undercharging.\n","But if ours is mispricing itself, are others? It's reasonable to expect that some resorts will be \"overpriced\" and some \"underpriced.\" Or if resorts are pretty good at pricing strategies, it could be that our model is simply lacking some key data? Certainly we know nothing about operating costs, for example, and they would surely help."]},{"cell_type":"markdown","metadata":{"id":"C-p69uVLXvBW"},"source":["## 5.8 Big Mountain Resort In Market Context"]},{"cell_type":"markdown","metadata":{"id":"fosXfx2HXvBW"},"source":["Features that came up as important in the modeling (not just our final, random forest model) included:\n","* vertical_drop\n","* Snow Making_ac\n","* total_chairs\n","* fastQuads\n","* Runs\n","* LongestRun_mi\n","* trams\n","* SkiableTerrain_ac"]},{"cell_type":"markdown","metadata":{"id":"wPU523JLXvBW"},"source":["A handy glossary of skiing terms can be found on the [ski.com](https://www.ski.com/ski-glossary) site. Some potentially relevant contextual information is that vertical drop, although nominally the height difference from the summit to the base, is generally taken from the highest [_lift-served_](http://verticalfeet.com/) point."]},{"cell_type":"markdown","metadata":{"id":"QnNgg9hpXvBW"},"source":["It's often useful to define custom functions for visualizing data in meaningful ways. The function below takes a feature name as an input and plots a histogram of the values of that feature. It then marks where Big Mountain sits in the distribution by marking Big Mountain's value with a vertical line using `matplotlib`'s [axvline](https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.axvline.html) function. It also performs a little cleaning up of missing values and adds descriptive labels and a title."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"rJa5shoGXvBW"},"outputs":[],"source":["#Code task 1#\n","#Add code to the `plot_compare` function that displays a vertical, dashed line\n","#on the histogram to indicate Big Mountain's position in the distribution\n","#Hint: plt.axvline() plots a vertical line, its position for 'feature1'\n","#would be `big_mountain['feature1'].values, we'd like a red line, which can be\n","#specified with c='r', a dashed linestyle is produced by ls='--',\n","#and it's nice to give it a slightly reduced alpha value, such as 0.8.\n","#Don't forget to give it a useful label (e.g. 'Big Mountain') so it's listed\n","#in the legend.\n","import matplotlib.pyplot as plt # Import the matplotlib.pyplot module\n","\n","def plot_compare(feat_name, description, state=None, figsize=(10, 5)):\n"," \"\"\"Graphically compare distributions of features.\n","\n"," Plot histogram of values for all resorts and reference line to mark\n"," Big Mountain's position.\n","\n"," Arguments:\n"," feat_name - the feature column name in the data\n"," description - text description of the feature\n"," state - select a specific state (None for all states)\n"," figsize - (optional) figure size\n"," \"\"\"\n","\n"," plt.subplots(figsize=figsize)\n"," # quirk that hist sometimes objects to NaNs, sometimes doesn't\n"," # filtering only for finite values tidies this up\n"," if state is None:\n"," ski_x = ski_data[feat_name]\n"," else:\n"," ski_x = ski_data.loc[ski_data.state == state, feat_name]\n"," ski_x = ski_x[np.isfinite(ski_x)]\n"," plt.hist(ski_x, bins=30)\n"," plt.axvline(x=big_mountain[feat_name].values, c='r', ls='--', alpha=0.8, label='Big Mountain')\n"," plt.xlabel(description)\n"," plt.ylabel('frequency')\n"," plt.title(description + ' distribution for resorts in market share')\n"," plt.legend()"]},{"cell_type":"markdown","metadata":{"id":"-y5Q3dEtXvBX"},"source":["### 5.8.1 Ticket price"]},{"cell_type":"markdown","metadata":{"id":"hNQZFl7wXvBX"},"source":["Look at where Big Mountain sits overall amongst all resorts for price and for just other resorts in Montana."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"Dk4ghqFFXvBX","executionInfo":{"status":"ok","timestamp":1721138838597,"user_tz":240,"elapsed":382,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"f425b0e8-38ad-48ee-9071-b624dc2fb85c"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('AdultWeekend', 'Adult weekend ticket price ($)')"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"B2Xbqot-XvBX","executionInfo":{"status":"ok","timestamp":1721138871501,"user_tz":240,"elapsed":485,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"96d41bd8-9d70-4d45-f895-e9999f7a306f"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('AdultWeekend', 'Adult weekend ticket price ($) - Montana only', state='Montana')"]},{"cell_type":"markdown","metadata":{"id":"6VpYubEYXvBX"},"source":["### 5.8.2 Vertical drop"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"clXnvyC3XvBY","executionInfo":{"status":"ok","timestamp":1721138882254,"user_tz":240,"elapsed":506,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"c548c394-25a2-49df-9d13-86ecf336d452"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('vertical_drop', 'Vertical drop (feet)')"]},{"cell_type":"markdown","metadata":{"id":"8owrkoAsXvBY"},"source":["Big Mountain is doing well for vertical drop, but there are still quite a few resorts with a greater drop."]},{"cell_type":"markdown","metadata":{"id":"P89un3eLXvBd"},"source":["### 5.8.3 Snow making area"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"noE5PISxXvBd","executionInfo":{"status":"ok","timestamp":1721138899684,"user_tz":240,"elapsed":420,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"f6438436-fd2e-4afb-8a10-83dcfe8497c4"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAA1oAAAHWCAYAAABuT/gUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB6s0lEQVR4nO3dd1QU19sH8O9SdkGqIFVREIm9F8RuRLHE2MUSxRI19t6Sny1RsZfYiImiJhpb1FhJFLtRLLEXRMUuoCIgSt/7/sHLxJEi6OACfj/nzDmzd+7ceabs7D47M3dVQggBIiIiIiIiUoyergMgIiIiIiIqaJhoERERERERKYyJFhERERERkcKYaBERERERESmMiRYREREREZHCmGgREREREREpjIkWERERERGRwphoERERERERKYyJFhERERERkcKYaBHlY87OzujVq1eWde7evQuVSoV58+Z9nKDonZydnfHFF1/oOox3GjRoEJo2barrMD6In58fihcvjoSEhA9qp1GjRmjUqJH0Ou19tWbNmg8LMBvWrFkDlUqFu3fvSmUf8xg6fPgwVCoVDh8+/FGW97a5c+eiZMmS0NfXR5UqVXQSw6egV69ecHZ21nUYisrrn39p762tW7fqOhTKJUy06J2WL18OlUoFd3d3XYdCRB9JaGgofvnlF3z77be6DuWD9OrVC4mJifjpp590HQqA1PPpx0jO3kdejO3vv//GuHHjULduXfj7+2PmzJm6DilPePz4MaZOnYoLFy7oOpRPyoYNG7Bo0SJdh0H5iIGuA6C8b/369XB2dsbp06dx69YtlCpVStchEVEuW7x4MVxcXNC4cWNdh/JBjIyM4OPjgwULFmDo0KFQqVSKtFuiRAnExcXB0NAwR/MtX74cRYoUeeeV6Df16NEDXbp0gUajyWGUOZNZbA0aNEBcXBzUanWuLj8jBw8ehJ6eHlatWqWT5edVjx8/xrRp0+Ds7KzYVb6ff/4ZWq1WkbYKqg0bNuDKlSsYMWKErkOhfIJXtChLoaGh+Oeff7BgwQLY2Nhg/fr12ZovOTkZiYmJuRxd/vXq1Stdh0AF2IceX0lJSVi/fj06d+6sUETZo9VqER8fr3i7nTt3xr1793Do0CHF2lSpVDAyMoK+vr5ibb4tbT/q6+vDyMhIsSQxp/T09GBkZAQ9vY//lSEiIgLGxsaKJVlCCMTFxeVontevXyuybCXk5meroaFhrifzH0tunUvyK37n0B0mWpSl9evXo3DhwmjVqhU6duyYYaL15j3QixYtgqurKzQaDa5duwYAuHHjBjp27AgrKysYGRmhRo0a2Llzp6yNyMhIjBkzBhUrVoSpqSnMzc3RokULXLx4Mdux/vbbb6hVqxYKFSqEwoULo0GDBvj7779ldZYvX47y5ctDo9HA0dERgwcPRlRUlDR9yJAhMDU1zfCDtWvXrrC3t0dKSopUtm/fPtSvXx8mJiYwMzNDq1atcPXqVdl8vXr1gqmpKW7fvo2WLVvCzMwM3bt3B5D6YbBo0SKUL18eRkZGsLOzw4ABA/DixQtZG0IITJ8+HcWKFUOhQoXQuHHjdMvJjoULF6JEiRIwNjZGw4YNceXKFWmav78/VCoVzp8/n26+mTNnQl9fH48ePcq07ZcvX2LEiBFwdnaGRqOBra0tmjZtin///Veq06hRI1SoUAHXrl1D48aNUahQIRQtWhRz5sxJ115ERAT69u0LOzs7GBkZoXLlyli7dq2sTrVq1dC+fXtZWcWKFaFSqXDp0iWpbNOmTVCpVLh+/Xqm8afdK79582ZMmzYNRYsWhZmZGTp27Ijo6GgkJCRgxIgRsLW1hampKXr37p3uuR9/f398/vnnsLW1hUajQbly5bBixYpMl/mmtWvXwsDAAGPHjpXKgoKC0Lx5c1hYWKBQoUJo2LAhTpw4IZtv6tSpUKlUuHbtGrp164bChQujXr16AICwsDD07t0bxYoVg0ajgYODA9q0aSN71icjx48fx7Nnz+Dp6SkrT0xMxOTJk1G9enVYWFjAxMQE9evXzzCB0Wq1WLx4MSpWrAgjIyPY2NigefPmOHv2rFRHpVJhyJAhWL9+vfS+DAgIAAA8evQIffr0gZ2dHTQaDcqXL4/Vq1enW86SJUtQvnx56X1fo0YNbNiwQVanevXqsLKywp9//pnleqdZuXIlXF1dYWxsjFq1auHYsWPp6mT0jNa7trezszOuXr2KI0eOQKVSQaVSSc99pT2HdeTIEQwaNAi2trYoVqyYbFpG++3vv/9GlSpVYGRkhHLlymHbtm2y6WnHx9vebjOr2DJ7RmvLli2oXr06jI2NUaRIEXz11VfpzhFp579Hjx6hbdu2MDU1hY2NDcaMGSM7l2ZEpVLB398fr169kmJK297Jycn44YcfpM8bZ2dnfPvtt+nek2nPsv3111+oUaMGjI2Ns7yNNO0cde7cOTRo0ACFChWSbp9NSEjAlClTUKpUKWg0Gjg5OWHcuHHplrl//37Uq1cPlpaWMDU1RenSpdPdgpud81tmn63Lly9HzZo1AQC9e/dOt21CQkLQoUMH2Nvbw8jICMWKFUOXLl0QHR2d5fZ++xmtN5ef9p7QaDSoWbMmzpw5k2VbwH/H2PHjxzFs2DDY2NjA0tISAwYMQGJiIqKiotCzZ08ULlwYhQsXxrhx4yCEkLUxb9481KlTB9bW1jA2Nkb16tUzfJ4pq3PJ24QQ6N+/P9Rqtez98ttvv0nHs5WVFbp06YIHDx5I0xs1aoQ9e/bg3r170jZ/1zNt2TkWgNTz5YwZM1CsWDEYGRmhSZMmuHXrlqzOsWPH0KlTJxQvXlw6/kaOHJnuhwMlvnOQcnjrIGVp/fr1aN++PdRqNbp27YoVK1bgzJkz0kn+Tf7+/oiPj0f//v2h0WhgZWWFq1evom7duihatCgmTJgAExMTbN68GW3btsUff/yBdu3aAQDu3LmDHTt2oFOnTnBxcUF4eDh++uknNGzYENeuXYOjo2OWcU6bNg1Tp05FnTp18P3330OtViMoKAgHDx5Es2bNAKR+4Zg2bRo8PT0xcOBABAcHS+tz4sQJGBoawtvbG8uWLcOePXvQqVMnqf3Xr19j165d6NWrl/QL9q+//gofHx94eXlh9uzZeP36NVasWIF69erh/PnzshNwcnIyvLy8UK9ePcybNw+FChUCAAwYMABr1qxB7969MWzYMISGhmLp0qU4f/68FBMATJ48GdOnT0fLli3RsmVL/Pvvv2jWrFmOftlct24dXr58icGDByM+Ph6LFy/G559/jsuXL8POzg4dO3bE4MGDsX79elStWjXdcdCoUSMULVo00/a/+eYbbN26FUOGDEG5cuXw/PlzHD9+HNevX0e1atWkei9evEDz5s3Rvn17dO7cGVu3bsX48eNRsWJFtGjRAgAQFxeHRo0a4datWxgyZAhcXFywZcsW9OrVC1FRURg+fDgAoH79+vj999+ltiMjI3H16lXo6enh2LFjqFSpEoDUDygbGxuULVv2ndvJ19cXxsbGmDBhAm7duoUlS5bA0NAQenp6ePHiBaZOnYpTp05hzZo1cHFxweTJk6V5V6xYgfLly+PLL7+EgYEBdu3ahUGDBkGr1WLw4MGZLnPlypX45ptv8O2332L69OkAUm+ZatGiBapXr44pU6ZAT09PSuSOHTuGWrVqydro1KkT3NzcMHPmTOnLSocOHXD16lUMHToUzs7OiIiIwP79+3H//v0svyD8888/UKlU6Y6DmJgY/PLLL+jatSv69euHly9fYtWqVfDy8sLp06dltzD17dsXa9asQYsWLfD1118jOTkZx44dw6lTp1CjRg2p3sGDB7F582YMGTIERYoUgbOzM8LDw1G7dm3py5ONjQ327duHvn37IiYmRrpt5+eff8awYcPQsWNHDB8+HPHx8bh06RKCgoLQrVs3WezVqlVLl6RmZNWqVRgwYADq1KmDESNG4M6dO/jyyy9hZWUFJyenLOd91/ZetGgRhg4dClNTU3z33XcAADs7O1kbgwYNgo2NDSZPnvzOX6FDQkLg7e2Nb775Bj4+PvD390enTp0QEBCQ405MshPbm9LOWzVr1oSvry/Cw8OxePFinDhxAufPn4elpaVUNyUlBV5eXnB3d8e8efNw4MABzJ8/H66urhg4cGCmy/j111+xcuVKnD59Gr/88gsAoE6dOgCAr7/+GmvXrkXHjh0xevRoBAUFwdfXF9evX8f27dtl7QQHB6Nr164YMGAA+vXrh9KlS2e5LZ4/f44WLVqgS5cu+Oqrr2BnZwetVosvv/wSx48fR//+/VG2bFlcvnwZCxcuxM2bN7Fjxw4AwNWrV/HFF1+gUqVK+P7776HRaHDr1i3ZsZfd81uatz9b27Vrh5cvX2Ly5Mno378/6tevL22bxMREeHl5ISEhAUOHDoW9vT0ePXqE3bt3IyoqChYWFlmue0Y2bNiAly9fYsCAAVCpVJgzZw7at2+PO3fuZOvW2bQ4pk2bhlOnTmHlypWwtLTEP//8g+LFi2PmzJnYu3cv5s6diwoVKqBnz57SvIsXL8aXX36J7t27IzExERs3bkSnTp2we/dutGrVSracjM4lb0tJSUGfPn2wadMmbN++XWpjxowZmDRpEjp37oyvv/4aT58+xZIlS9CgQQPpeP7uu+8QHR2Nhw8fYuHChQAAU1PTTNc7O8dCmlmzZkFPTw9jxoxBdHQ05syZg+7duyMoKEiqs2XLFrx+/RoDBw6EtbU1Tp8+jSVLluDhw4fYsmWLrL0P/c5BChJEmTh79qwAIPbv3y+EEEKr1YpixYqJ4cOHy+qFhoYKAMLc3FxERETIpjVp0kRUrFhRxMfHS2VarVbUqVNHuLm5SWXx8fEiJSUlXbsajUZ8//33WcYZEhIi9PT0RLt27dK1odVqhRBCRERECLVaLZo1ayars3TpUgFArF69WqpftGhR0aFDB1k7mzdvFgDE0aNHhRBCvHz5UlhaWop+/frJ6oWFhQkLCwtZuY+PjwAgJkyYIKt77NgxAUCsX79eVh4QECArT4u9VatW0voIIcS3334rAAgfH58st0/a/jE2NhYPHz6UyoOCggQAMXLkSKmsa9euwtHRUbaN/v33XwFA+Pv7Z7kcCwsLMXjw4CzrNGzYUAAQ69atk8oSEhKEvb29bJsvWrRIABC//fabVJaYmCg8PDyEqampiImJEUIIsWXLFgFAXLt2TQghxM6dO4VGoxFffvml8Pb2luatVKmSaNeuXZaxHTp0SAAQFSpUEImJiVJ5165dhUqlEi1atJDV9/DwECVKlJCVvX79Ol27Xl5eomTJkrKyEiVKiFatWgkhhFi8eLFQqVTihx9+kKZrtVrh5uYmvLy8ZPv89evXwsXFRTRt2lQqmzJligAgunbtKlvGixcvBAAxd+7cLNc7I1999ZWwtrZOV56cnCwSEhLSLcfOzk706dNHKjt48KAAIIYNG5aujTfXB4DQ09MTV69eldXp27evcHBwEM+ePZOVd+nSRVhYWEjbuU2bNqJ8+fLZWqf+/fsLY2PjLOskJiYKW1tbUaVKFdl6rly5UgAQDRs2lMrS3ldp74vsbu/y5cvL2knj7+8vAIh69eqJ5OTkDKeFhoZKZSVKlBAAxB9//CGVRUdHCwcHB1G1alWpLO34yGx5b7aZWWxp741Dhw4JIf7bThUqVBBxcXFSvd27dwsAYvLkyVJZ2vnv7fN41apVRfXq1dMt620+Pj7CxMREVnbhwgUBQHz99dey8jFjxggA4uDBg1JZ2nYKCAh457KE+O8c5efnJyv/9ddfhZ6enjh27Jis3M/PTwAQJ06cEEIIsXDhQgFAPH36NNNlZPf8ltVn65kzZzI8L58/f14AEFu2bMnW+r7Jx8dHdk5LW761tbWIjIyUyv/8808BQOzatSvL9tKOsbfPYx4eHkKlUolvvvlGKktOThbFihVLd/y9fU5NTEwUFSpUEJ9//rmsPLNzSdo6zJ07VyQlJQlvb29hbGws/vrrL6nO3bt3hb6+vpgxY4Zs3suXLwsDAwNZeatWrdKd9zOTnWMh7b1VtmxZ2Tln8eLFAoC4fPmyVJbR54uvr69QqVTi3r17UtmHfucgZfHWQcrU+vXrYWdnJz0Mr1Kp4O3tjY0bN2Z4y0eHDh1gY2MjvY6MjMTBgwfRuXNnvHz5Es+ePcOzZ8/w/PlzeHl5ISQkRLrNRKPRSPf/p6Sk4Pnz59Jl9jdvPcvIjh07oNVqMXny5HTPEKTdMnPgwAEkJiZixIgRsjr9+vWDubk59uzZI9Xv1KkT9u7di9jYWKnepk2bULRoUemWrP379yMqKgpdu3aV1uvZs2fQ19eHu7t7hrdSvf3L7ZYtW2BhYYGmTZvK2qhevTpMTU2lNtJif/tB/pw+jNu2bVvZFalatWrB3d0de/fulcp69uyJx48fy+Jfv349jI2N0aFDhyzbt7S0RFBQEB4/fpxlPVNTU3z11VfSa7VajVq1auHOnTtS2d69e2Fvb4+uXbtKZYaGhhg2bBhiY2Nx5MgRAJB+yT169CiA1CtXNWvWRNOmTaXbvaKionDlyhWp7rv07NlT9queu7s7hBDo06ePrJ67uzsePHiA5ORkqczY2Fgaj46OxrNnz9CwYUPcuXMnw9t25syZg+HDh2P27Nn43//+J5VfuHABISEh6NatG54/fy4dG69evUKTJk1w9OjRdA+tf/PNN7LXac+1HD58OMe3hTx//hyFCxdOV66vry89K6PVahEZGYnk5GTUqFFD9j79448/oFKpMGXKlHRtvH0bW8OGDVGuXDnptRACf/zxB1q3bg0hhOy94eXlhejoaGlZlpaWePjwYbZuYypcuDDi4uKyfN7m7NmziIiIwDfffCN7JqhXr17vvBLwIdv7Tf369cv2c1+Ojo7SXQEAYG5ujp49e+L8+fMICwt77xjeJW07DRo0CEZGRlJ5q1atUKZMGel8+qa3j8/69evL3vM5kXbOGjVqlKx89OjRAJBu+S4uLvDy8sp2+xqNBr1795aVbdmyBWXLlkWZMmVkx+Tnn38OANI5M+1K3p9//plpxxLZPb+lefuzNStpx+lff/2l2LNl3t7esvNB2rk0u/uvb9++svd92jm1b9++Upm+vj5q1KiRrs03z6kvXrxAdHQ06tevn+H3grfPJW9KTEyUroTt3btXutMFALZt2watVovOnTvL9q29vT3c3Nze+9nO7BwLaXr37i0752S0jd/cFq9evcKzZ89Qp04dCCEyvOX/fb9zkLKYaFGGUlJSsHHjRjRu3BihoaG4desWbt26BXd3d4SHhyMwMDDdPC4uLrLXt27dghACkyZNgo2NjWxI+wIWEREBIPVL28KFC+Hm5gaNRoMiRYrAxsYGly5deud95bdv34aenl6mJ1gAuHfvHgCku2VErVajZMmS0nQg9UMlLi5Oeo4sNjYWe/fuRadOnaQPi5CQEADA559/nm7d/v77b2m90hgYGEjPW6QJCQlBdHQ0bG1t07URGxsrtZEWm5ubm2x+GxubDL8MZ+bt+QHgs88+kz330bRpUzg4OEjP4mm1Wvz+++9o06YNzMzMsmx/zpw5uHLlCpycnFCrVi1MnTo1ww/iYsWKpfuyXbhwYdmX03v37sHNzS1d4px261/aNrGzs4Obm5uUVB07dgz169dHgwYN8PjxY9y5cwcnTpyAVqvNdqJVvHhx2eu0Ly5v3zZmYWEBrVYrOz5PnDgBT09PmJiYwNLSEjY2NtL9+G8fx0eOHMH48eMxfvx42XNZwH/Hl4+PT7pj45dffkFCQkK69t5+/2k0GsyePRv79u2DnZ0dGjRogDlz5mT7C7h461mJNGvXrkWlSpVgZGQEa2tr2NjYYM+ePbJ4bt++DUdHR1hZWb1zOW/H/fTpU0RFRWHlypXp1j3ty2/ae2P8+PEwNTVFrVq14ObmhsGDB2d6e2Da+mTVoURm7zVDQ0OULFkyy/X40O2d5u3tkZVSpUqlW5/PPvsMAN75HN6HyOx8CgBlypSRnU8BSM/ovent93xOl6+np5euB1x7e3tYWlqmW35OtikAFC1aNF3nGyEhIbh69Wq6YzJte6cdk97e3qhbty6+/vpr2NnZoUuXLti8ebPsi3Z2z2/vE7+LiwtGjRqFX375BUWKFIGXlxeWLVv2zs/RrLx9Tkz73Mnu/svJOfXtNnfv3o3atWvDyMgIVlZWsLGxwYoVKzJcn6y2k6+vL3bs2IGtW7fK/g8PSN23Qgi4ubml27/Xr19P93meXdk5FtJkZxvfv38fvXr1gpWVlfSsY8OGDQGk/3z5kO8cpCw+o0UZOnjwIJ48eYKNGzdi48aN6aavX79e9osQIP+1BYB0MhkzZkymvyamfVDOnDkTkyZNQp8+ffDDDz/AysoKenp6GDFixEfvbrZ27dpwdnbG5s2b0a1bN+zatQtxcXHw9vaW6qTF9Ouvv8Le3j5dGwYG8rfWm1fs3mzD1tY2054cs/sLppL09fXRrVs3/Pzzz1i+fDlOnDiBx48fy65AZaZz586oX78+tm/fjr///htz587F7NmzsW3bNunZq7RlZCSzL/bvUq9ePQQGBiIuLg7nzp3D5MmTUaFCBVhaWuLYsWO4fv06TE1N0z1vlJnM4ntX3Ldv30aTJk1QpkwZLFiwAE5OTlCr1di7dy8WLlyY7jguX748oqKi8Ouvv2LAgAGyLwlpdefOnZtp181vPxvw9vsPSL3q2bp1a+zYsQN//fUXJk2aBF9fXxw8eDDL7WFtbZ3hl6jffvsNvXr1Qtu2bTF27FjY2tpCX18fvr6+uH37dqbtZSWz88ZXX30FHx+fDOdJe/aubNmyCA4Oxu7duxEQEIA//vgDy5cvx+TJkzFt2jTZPC9evEChQoUy3E5Ked/t/Sal48sssXxXRxRKyq2eGbPbC2NOt2lG9bVaLSpWrIgFCxZkOE9a0mBsbIyjR4/i0KFD2LNnDwICArBp0yZ8/vnn+Pvvv99rW+Q0/vnz56NXr174888/8ffff2PYsGHw9fXFqVOn0n35zo4PPWfn5Jz6ZpvHjh3Dl19+iQYNGmD58uVwcHCAoaEh/P3903V4A2S9nby8vBAQEIA5c+agUaNGsiuxWq0WKpUK+/btyzCmrJ7DykpOjoV3beOUlBQ0bdoUkZGRGD9+PMqUKQMTExM8evQIvXr1Svf5kl++c3wKmGhRhtavXw9bW1ssW7Ys3bRt27Zh+/bt8PPzy/LElvYLsKGhYbrey962detWNG7cGKtWrZKVR0VFoUiRIlnO6+rqCq1Wi2vXrmX6pbREiRIAUh+KfvOX6cTERISGhqaLr3Pnzli8eDFiYmKwadMmODs7o3bt2rJlAoCtre071y2ruA8cOIC6detmuR3TYg8JCZHF/vTp0xz9Ipx2leRNN2/eTPfAcM+ePTF//nzs2rUL+/btg42NTbZvu3FwcMCgQYMwaNAgREREoFq1apgxY4Ys0cqOEiVK4NKlS9BqtbIPixs3bkjT09SvXx/+/v7SLa116tSBnp4e6tWrJyVaderUydVuuAFg165dSEhIwM6dO2W/TmZ2O0aRIkWwdetW1KtXD02aNMHx48elTl/Sji9zc/P3Pr7SuLq6YvTo0Rg9ejRCQkJQpUoVzJ8/H7/99lum85QpUwbr169HdHS07Ja5rVu3omTJkti2bZvsS+7btwi6urrir7/+QmRkZLauar3JxsYGZmZmSElJyda6m5iYwNvbG97e3khMTET79u0xY8YMTJw4UfZlKjQ09J2dobz5Xku7JQxI7e4+NDQUlStXfmc879reSnbRnnbXwJtt3rx5EwCk93XaL+NRUVGyDirevmqSk9jePJ++uZ3Syt58f+aGEiVKQKvVIiQkRLZPw8PDERUVlSvLd3V1xcWLF9GkSZN3bic9PT00adIETZo0wYIFCzBz5kx89913OHToEDw9PXN0fsvMu2KoWLEiKlasiP/973/4559/ULduXfj5+Umd7eQHf/zxB4yMjPDXX3/Jup339/fPcVu1a9fGN998gy+++AKdOnXC9u3bpR9EXV1dIYSAi4uLdIUyMzl9/77rWMiuy5cv4+bNm1i7dq2ss5D9+/dnu43sfucgZfHWQUonLi4O27ZtwxdffIGOHTumG4YMGYKXL1+m66L9bba2tmjUqBF++uknPHnyJN30p0+fSuP6+vrpfh3bsmVLlt2Jp2nbti309PTw/fffp/tVJ61NT09PqNVq/Pjjj7LlrFq1CtHR0el6L/L29kZCQgLWrl2LgICAdP8n5OXlBXNzc8ycORNJSUlZrltmOnfujJSUFPzwww/ppiUnJ0vdznt6esLQ0BBLliyRxZ7Tf6ffsWOHbHuePn0aQUFB6ZKgSpUqoVKlSvjll1/wxx9/oEuXLumu0L0tJSUl3a0Ltra2cHR0TNf1cXa0bNkSYWFh2LRpk1SWnJyMJUuWwNTUVLpdAvjvXvbZs2ejUqVKUmJQv359BAYG4uzZs9m+bfBDpCVyb+6j6OjoLL8UFCtWDAcOHEBcXByaNm2K58+fA0jtjtzV1RXz5s2TPSuYJjvH1+vXr9P9j4yrqyvMzMzeuU88PDwghMC5c+dk5RmtY1BQEE6ePCmr16FDBwgh0l1VenvejOjr66NDhw74448/ZH8/kObNdU/bXmnUajXKlSsHIUS69+W///4r9ViXmRo1asDGxgZ+fn6yHj3XrFkj+xuIjGR3e5uYmLyzrex6/PixrIe9mJgYrFu3DlWqVJGutKcl7WnPMQKpz3e83ZV4TmKrUaMGbG1t4efnJ1u3ffv24fr16+nOp0pr2bIlgPTnwLSrTbmx/M6dO+PRo0f4+eef002Li4uTeoiMjIxMNz3tB8C0bZWT81tmTExMACDd/oqJiZE9NwqkJl16enrvdS7WJX19fahUKtnV17t370o9POaUp6cnNm7ciICAAPTo0UP6vtC+fXvo6+tj2rRp6c5PQgjZecbExCTbt2Fm51jIrozOvUIILF68ONttZPc7BymLV7QonZ07d+Lly5f48ssvM5xeu3Zt6c+L37ydLiPLli1DvXr1ULFiRfTr1w8lS5ZEeHg4Tp48iYcPH0r/k/XFF1/g+++/R+/evVGnTh1cvnwZ69evf+dzEUDq7YffffcdfvjhB9SvXx/t27eHRqPBmTNn4OjoCF9fX9jY2GDixImYNm0amjdvji+//BLBwcHS/5G8fWtctWrVpHYTEhLSrae5uTlWrFiBHj16oFq1aujSpQtsbGxw//597NmzB3Xr1sXSpUuzjLthw4YYMGAAfH19ceHCBTRr1gyGhoYICQnBli1bsHjxYnTs2FH6zxlfX1988cUXaNmyJc6fP499+/a982rf29upXr16GDhwIBISErBo0SJYW1tj3Lhx6er27NkTY8aMAYBs3Tb48uVLFCtWDB07dkTlypVhamqKAwcO4MyZM5g/f362Y0zTv39//PTTT+jVqxfOnTsHZ2dnbN26FSdOnMCiRYtkz4uVKlUK9vb2CA4OxtChQ6XyBg0aYPz48QDwURKtZs2aQa1Wo3Xr1hgwYABiY2Px888/w9bWNsMfGt6M/++//0ajRo3g5eWFgwcPwtzcHL/88gtatGiB8uXLo3fv3ihatCgePXqEQ4cOwdzcHLt27coynps3b6JJkybo3LkzypUrBwMDA2zfvh3h4eHo0qVLlvPWq1cP1tbWOHDggOyKxRdffIFt27ahXbt2aNWqFUJDQ+Hn54dy5crJEsLGjRujR48e+PHHHxESEoLmzZtDq9Xi2LFjaNy4MYYMGZLl8mfNmoVDhw7B3d0d/fr1Q7ly5RAZGYl///0XBw4ckL7ANGvWDPb29qhbty7s7Oxw/fp1LF26FK1atZIdI+fOnUNkZCTatGmT5XINDQ0xffp0DBgwAJ9//jm8vb0RGhoKf3//d56Lsru9q1evjhUrVmD69OkoVaoUbG1t010Vyq7PPvsMffv2xZkzZ2BnZ4fVq1cjPDxcltw3a9YMxYsXR9++fTF27Fjo6+tj9erV0vnqTdmNzdDQELNnz0bv3r3RsGFDdO3aVere3dnZGSNHjnyv9cmuypUrw8fHBytXrkRUVBQaNmyI06dPY+3atWjbtq3UgZOSevTogc2bN+Obb77BoUOHULduXaSkpODGjRvYvHmz9D9d33//PY4ePYpWrVqhRIkSiIiIwPLly1GsWDGpM6WcnN8y4+rqCktLS/j5+cHMzAwmJiZwd3fHxYsXMWTIEHTq1AmfffYZkpOT8euvv0o/YOQnrVq1woIFC9C8eXN069YNERERWLZsGUqVKiX7n8ScaNu2Lfz9/dGzZ0+Ym5vjp59+gqurK6ZPn46JEyfi7t27aNu2LczMzBAaGort27ejf//+0udh9erVsWnTJowaNQo1a9aEqakpWrduneGysnMsZFeZMmXg6uqKMWPG4NGjRzA3N8cff/yRo7tasvudgxT2Mbo2pPyldevWwsjISLx69SrTOr169RKGhobi2bNnsu5TM3L79m3Rs2dPYW9vLwwNDUXRokXFF198IbZu3SrViY+PF6NHjxYODg7C2NhY1K1bV5w8eVI0bNgww+6GM7J69WpRtWpVodFoROHChUXDhg2lrunTLF26VJQpU0YYGhoKOzs7MXDgQPHixYsM2/vuu+8EAFGqVKlMl3no0CHh5eUlLCwshJGRkXB1dRW9evUSZ8+elepk1D3xm1auXCmqV68ujI2NhZmZmahYsaIYN26cePz4sVQnJSVFTJs2Tdo+jRo1EleuXBElSpTIdvfuc+fOFfPnzxdOTk5Co9GI+vXri4sXL2Y4z5MnT4S+vr747LPPsmw7TUJCghg7dqyoXLmyMDMzEyYmJqJy5cpi+fLlsnoNGzbMsDvut7sVFkKI8PBw0bt3b1GkSBGhVqtFxYoVM+1ivlOnTgKA2LRpk1SWmJgoChUqJNRqtawL6sykdbP7drfIaV0UnzlzRlae1m32m1337ty5U1SqVEkYGRkJZ2dnMXv2bLF69eoMu+ZO6949TVBQkDAzMxMNGjSQuvE9f/68aN++vbC2thYajUaUKFFCdO7cWQQGBmYZhxBCPHv2TAwePFiUKVNGmJiYCAsLC+Hu7i42b978zm0hhBDDhg1Ld+xrtVoxc+ZMUaJECaHRaETVqlXF7t27M9x/ycnJYu7cuaJMmTJCrVYLGxsb0aJFC3Hu3DmpDoBM/xIgPDxcDB48WDg5OQlDQ0Nhb28vmjRpIlauXCnV+emnn0SDBg2k7ePq6irGjh0roqOjZW2NHz9eFC9eXNbFdFaWL18uXFxchEajETVq1BBHjx5Ndy56u3v37G7vsLAw0apVK2FmZibrMj6z4+zNaRkdQ3/99ZeoVKmS0Gg0okyZMhl2633u3Dnh7u4u1Gq1KF68uFiwYEGGbWYW29vdu6fZtGmTdM61srIS3bt3l/2FhBCZn/8y63b+bZnNn5SUJKZNmyZcXFyEoaGhcHJyEhMnTpT9lcib2ym7MjtHCZF6Tpk9e7YoX7689DlTvXp1MW3aNOmYCwwMFG3atBGOjo5CrVYLR0dH0bVrV3Hz5k1ZW9k5v73rs/XPP/8U5cqVEwYGBtKxeOfOHdGnTx/h6uoqjIyMhJWVlWjcuLE4cODAO9c9s+7dM1o+ADFlypQs28vJuTNt+W/v61WrVgk3Nzfp+Pb398/w2MnsXJLZOixfvlwAEGPGjJHK/vjjD1GvXj1hYmIiTExMRJkyZcTgwYNFcHCwVCc2NlZ069ZNWFpaCgBZdvWenWMhs8+dt88vQghx7do14enpKUxNTUWRIkVEv379xMWLF9PVU+I7BylHJcR7PoFORAXWs2fP4ODggMmTJ2PSpEm6Dod04M6dOyhTpgz27duHJk2a6Dqc95aQkABnZ2dMmDAh3R/BEhER5SY+o0VE6axZswYpKSno0aOHrkMhHSlZsiT69u2LWbNm6TqUD+Lv7w9DQ8N0/+NERESU23hFi4gkBw8exLVr1zBp0iQ0btwY27Zt03VIRERERPkSEy0ikjRq1EjqCvi3335D0aJFdR0SERERUb7ERIuIiIiIiEhhOn1G6+jRo2jdujUcHR2hUqky/G+E69ev48svv4SFhQVMTExQs2ZNWZe08fHxGDx4MKytrWFqaooOHTogPDz8I64FERERERGRnE4TrVevXqFy5cpYtmxZhtNv376NevXqoUyZMjh8+DAuXbqESZMmwcjISKozcuRI7Nq1C1u2bMGRI0fw+PFjtG/f/mOtAhERERERUTp55tZBlUqF7du3o23btlJZly5dYGhoiF9//TXDeaKjo2FjY4MNGzZIf7J248YNlC1bFidPnkTt2rWztWytVovHjx/DzMwMKpXqg9eFiIiIiIjyJyEEXr58CUdHR+jpvf91KQMFY1KUVqvFnj17MG7cOHh5eeH8+fNwcXHBxIkTpWTs3LlzSEpKgqenpzRfmTJlULx48SwTrYSEBCQkJEivHz16hHLlyuXq+hARERERUf7x4MEDFCtW7L3nz7OJVkREBGJjYzFr1ixMnz4ds2fPRkBAANq3b49Dhw6hYcOGCAsLg1qthqWlpWxeOzs7hIWFZdq2r68vpk2blq78wYMHMDc3V3pViDKn1QJpzxTa2QEf8KsJEREREX24mJgYODk5wczM7IPaybOJllarBQC0adMGI0eOBABUqVIF//zzD/z8/NCwYcP3bnvixIkYNWqU9DptY5qbmzPRoo8rLg7o3j11/NgxwNhYt/EQEREREQB88CNFeTbRKlKkCAwMDNLd0le2bFkcP34cAGBvb4/ExERERUXJrmqFh4fD3t4+07Y1Gg00Gk2uxE1ERERERJRn71NSq9WoWbMmgoODZeU3b95EiRIlAADVq1eHoaEhAgMDpenBwcG4f/8+PDw8Pmq8REREREREaXR6RSs2Nha3bt2SXoeGhuLChQuwsrJC8eLFMXbsWHh7e6NBgwZo3LgxAgICsGvXLhw+fBgAYGFhgb59+2LUqFGwsrKCubk5hg4dCg8Pj2z3OEhERERERKQ0nSZaZ8+eRePGjaXXac9N+fj4YM2aNWjXrh38/Pzg6+uLYcOGoXTp0vjjjz9Qr149aZ6FCxdCT08PHTp0QEJCAry8vLB8+fKPvi5EREREpDtCCCQnJyMlJUXXoVAep6+vDwMDg1z/W6c88z9auhQTEwMLCwtER0ezMwz6uOLigPr1U8fZGQYREdF7SUxMxJMnT/D69Wtdh0L5RKFCheDg4AC1Wp1umlK5QZ7tDIOIiIiI6F20Wi1CQ0Ohr68PR0dHqNXqXL9SQfmXEAKJiYl4+vQpQkND4ebm9kF/SpwVJlpEuqSvD3Tq9N84ERER5UhiYiK0Wi2cnJxQqFAhXYdD+YCxsTEMDQ1x7949JCYmwsjIKFeWw0SLSJfUamD8eF1HQURElO/l1lUJKpg+xvHCI5KIiIiIiEhhvKJFpEtCAFFRqeOWlgDvKSciIiIqEHhFi0iX4uOBpk1Th/h4XUdDREREeczdu3ehUqlw4cIFXYeSpzg7O2PRokW6DiNLTLSIiIiIiHSgV69eUKlU0mBtbY3mzZvj0qVLUh0nJyc8efIEFSpU+KBlOTs7Q6VSYePGjemmlS9fHiqVCmvWrPmgZbwPlUqFHTt25Hi+M2fOoH///soHpCAmWkREREREOtK8eXM8efIET548QWBgIAwMDPDFF19I0/X19WFvbw8Dgw9/4sfJyQn+/v6yslOnTiEsLAwmJiYf3P7HZGNjk+d7mWSiRUREREQFU1xc5kNiYvbrJiRkr+570Gg0sLe3h729PapUqYIJEybgwYMHePr0KYCMbx3cuXMn3NzcYGRkhMaNG2Pt2rVQqVSISnvuOxPdu3fHkSNH8ODBA6ls9erV6N69e7pE7v79+2jTpg1MTU1hbm6Ozp07Izw8XJreq1cvtG3bVjbPiBEj0KhRI+l1o0aNMGzYMIwbNw5WVlawt7fH1KlTpenOzs4AgHbt2kGlUkmvb9++jTZt2sDOzg6mpqaoWbMmDhw4IFvW27cOqlQq/PLLL2jXrh0KFSoENzc37Ny5M8vtkduYaBERERFRwVS/fubD2LHyuk2bZl536FB53datM673gWJjY/Hbb7+hVKlSsLa2zrBOaGgoOnbsiLZt2+LixYsYMGAAvvvuu2y1b2dnBy8vL6xduxYA8Pr1a2zatAl9+vSR1dNqtWjTpg0iIyNx5MgR7N+/H3fu3IG3t3eO12nt2rUwMTFBUFAQ5syZg++//x779+8HkHr7HwD4+/vjyZMn0uvY2Fi0bNkSgYGBOH/+PJo3b47WrVvj/v37WS5r2rRp6Ny5My5duoSWLVuie/fuiIyMzHHMSmGvg3mQ84Q9udb23Vmtcq1tIiIiIsqZ3bt3w9TUFADw6tUrODg4YPfu3Zn+z9NPP/2E0qVLY+7cuQCA0qVL48qVK5gxY0a2ltenTx+MHj0a3333HbZu3QpXV1dUqVJFVicwMBCXL19GaGgonJycAADr1q1D+fLlcebMGdSsWTPb61epUiVMmTIFAODm5oalS5ciMDAQTZs2hY2NDQDA0tIS9vb20jyVK1dG5cqVpdc//PADtm/fjp07d2LIkCGZLqtXr17o2rUrAGDmzJn48ccfcfr0aTRv3jzb8SqJiRYRERERFUzHjmU+TV9f/vr/r7Jk6O2kZ9eu94/pLY0bN8aKFSsAAC9evMDy5cvRokULnD59GiVKlEhXPzg4OF2iU6tWrWwvr1WrVhgwYACOHj2K1atXp7uaBQDXr1+Hk5OTlGQBQLly5WBpaYnr16/nONF6k4ODAyIiIrKcJzY2FlOnTsWePXvw5MkTJCcnIy4u7p1XtN5clomJCczNzd+5rNzERItIl/T1gbQHXt8+4RMREdGHMTbWfd13MDExQalSpaTXv/zyCywsLPDzzz9j+vTpii0njYGBAXr06IEpU6YgKCgI27dvf6929PT0IISQlSUlJaWrZ2hoKHutUqmg1WqzbHvMmDHYv38/5s2bh1KlSsHY2BgdO3ZE4tvP1SmwrNzERItIl9Rq4I2HQomIiOjTplKpoKenh7hMOtcoXbo09u7dKytLe7Ypu/r06YN58+bB29sbhQsXTje9bNmyePDgAR48eCBd1bp27RqioqJQrlw5AKm9/l25ckU234ULF9IlO+9iaGiIlJQUWdmJEyfQq1cvtGvXDkDqFa67d+/mqN28gJ1hEBERERHpSEJCAsLCwhAWFobr169j6NChiI2NRevWrTOsP2DAANy4cQPjx4/HzZs3sXnzZun/r1QqVbaWWbZsWTx79ixdV+9pPD09UbFiRXTv3h3//vsvTp8+jZ49e6Jhw4aoUaMGAODzzz/H2bNnsW7dOoSEhGDKlCnpEq/scHZ2RmBgIMLCwvDixQsAqc9ybdu2DRcuXMDFixfRrVs3nV6Zel9MtIh0SYj/uoR96/I7ERERFXwBAQFwcHCAg4MD3N3dcebMGWzZskXWTfqbXFxcsHXrVmzbtg2VKlXCihUrpF4HNRpNtpdrbW0N40xugVSpVPjzzz9RuHBhNGjQAJ6enihZsiQ2bdok1fHy8sKkSZMwbtw41KxZEy9fvkTPnj2zv+L/b/78+di/fz+cnJxQtWpVAMCCBQtQuHBh1KlTB61bt4aXlxeqVauW47Z1TSXevrnyExQTEwMLCwtER0fD3Nxc1+Gw18FPSVzcf93BHjum6D3fREREn4L4+HiEhobCxcUFRkZGug5HJ2bMmAE/Pz/Z/2NR1rI6bpTKDfiMFhERERFRPrJ8+XLUrFkT1tbWOHHiBObOnZtlt+ekG0y0iIiIiIjykZCQEEyfPh2RkZEoXrw4Ro8ejYkTJ+o6LHoLEy0iIiIionxk4cKFWLhwoa7DoHdgZxhEREREREQKY6JFRERERPke+3ejnPgYxwsTLSIiIiLKt9L+IPf169c6joTyk7TjJad/sJwTfEaLSJf09YEmTf4bJyIiohzR19eHpaUlIiIiAACFChXK9h/30qdHCIHXr18jIiIClpaW0M/F719MtIh0Sa0GZs/WdRRERET5mr29PQBIyRbRu1haWkrHTW5hokVERERE+ZpKpYKDgwNsbW2RlJSk63AojzM0NMzVK1lpmGgRERERUYGgr6//Ub5AE2UHO8Mg0qW4OKBGjdQhLk7X0RARERGRQphoERERERERKYyJFhERERERkcKYaBERERERESmMiRYREREREZHCmGgREREREREpjIkWERERERGRwvg/WkS6pK8P1K373zgRERERFQhMtIh0Sa0GFi/WdRREREREpDDeOkhERERERKQwnSZaR48eRevWreHo6AiVSoUdO3ZkWvebb76BSqXCokWLZOWRkZHo3r07zM3NYWlpib59+yI2NjZ3AyciIiIiIsqCThOtV69eoXLlyli2bFmW9bZv345Tp07B0dEx3bTu3bvj6tWr2L9/P3bv3o2jR4+if//+uRUykbLi4oB69VKHuDhdR0NERERECtHpM1otWrRAixYtsqzz6NEjDB06FH/99RdatWolm3b9+nUEBATgzJkzqFGjBgBgyZIlaNmyJebNm5dhYkaU58TH6zoCIiIiIlJYnn5GS6vVokePHhg7dizKly+fbvrJkydhaWkpJVkA4OnpCT09PQQFBWXabkJCAmJiYmQDERERERGRUvJ0ojV79mwYGBhg2LBhGU4PCwuDra2trMzAwABWVlYICwvLtF1fX19YWFhIg5OTk6JxExERERHRpy3PJlrnzp3D4sWLsWbNGqhUKkXbnjhxIqKjo6XhwYMHirZPRERERESftjybaB07dgwREREoXrw4DAwMYGBggHv37mH06NFwdnYGANjb2yMiIkI2X3JyMiIjI2Fvb59p2xqNBubm5rKBiIiIiIhIKXn2D4t79OgBT09PWZmXlxd69OiB3r17AwA8PDwQFRWFc+fOoXr16gCAgwcPQqvVwt3d/aPHTEREREREBOg40YqNjcWtW7ek16Ghobhw4QKsrKxQvHhxWFtby+obGhrC3t4epUuXBgCULVsWzZs3R79+/eDn54ekpCQMGTIEXbp0YY+DlD/o6QHVqv03TkREREQFgk4TrbNnz6Jx48bS61GjRgEAfHx8sGbNmmy1sX79egwZMgRNmjSBnp4eOnTogB9//DE3wiVSnkYDrFyp6yiIiIiISGE6TbQaNWoEIUS269+9ezddmZWVFTZs2KBgVERERERERB+G9yoREREREREpjIkWkS7FxQGenqlDXJyuoyEiIiIiheTZXgeJPhlRUbqOgIiIiIgUxitaRERERERECmOiRUREREREpDAmWkRERERERApjokVERERERKQwJlpEREREREQKY6+DRLqkpweUK/ffOBEREREVCEy0iHRJowHWrdN1FERERESkMP6ETkREREREpDAmWkRERERERApjokWkS/HxQOvWqUN8vK6jISIiIiKF8BktIl0SAnjy5L9xIiIiIioQeEWLiIiIiIhIYUy0iIiIiIiIFMZEi4iIiIiISGFMtIiIiIiIiBTGRIuIiIiIiEhh7HWQSJdUKqBkyf/GiYiIiKhAYKJFpEtGRsDmzbqOgoiIiIgUxlsHiYiIiIiIFMZEi4iIiIiISGFMtIh0KT4e6Nw5dYiP13U0RERERKQQPqNFpEtCAHfu/DdORERERAUCr2gREREREREpjIkWERERERGRwphoERERERERKYyJFhERERERkcKYaBERERERESmMvQ4S6ZJKBTg4/DdORERERAUCEy0iXTIyAnbt0nUURERERKQw3jpIRERERESkMCZaRERERERECmOiRaRLCQlAz56pQ0KCrqMhIiIiIoXwGS0iXdJqgWvX/hsnIiIiogKBV7SIiIiIiIgUptNE6+jRo2jdujUcHR2hUqmwY8cOaVpSUhLGjx+PihUrwsTEBI6OjujZsyceP34sayMyMhLdu3eHubk5LC0t0bdvX8TGxn7kNSEiIiIiIvqPThOtV69eoXLlyli2bFm6aa9fv8a///6LSZMm4d9//8W2bdsQHByML7/8Ulave/fuuHr1Kvbv34/du3fj6NGj6N+//8daBSIiIiIionR0+oxWixYt0KJFiwynWVhYYP/+/bKypUuXolatWrh//z6KFy+O69evIyAgAGfOnEGNGjUAAEuWLEHLli0xb948ODo65vo6EBERERERvS1fPaMVHR0NlUoFS0tLAMDJkydhaWkpJVkA4OnpCT09PQQFBWXaTkJCAmJiYmQDERERERGRUvJNohUfH4/x48eja9euMDc3BwCEhYXB1tZWVs/AwABWVlYICwvLtC1fX19YWFhIg5OTU67GTpQlS8vUgYiIiIgKjHyRaCUlJaFz584QQmDFihUf3N7EiRMRHR0tDQ8ePFAgSqL3YGwMHDiQOhgb6zoaIiIiIlJInv8frbQk6969ezh48KB0NQsA7O3tERERIaufnJyMyMhI2NvbZ9qmRqOBRqPJtZiJiIiIiOjTlqevaKUlWSEhIThw4ACsra1l0z08PBAVFYVz585JZQcPHoRWq4W7u/vHDpeIiIiIiAiAjq9oxcbG4tatW9Lr0NBQXLhwAVZWVnBwcEDHjh3x77//Yvfu3UhJSZGeu7KysoJarUbZsmXRvHlz9OvXD35+fkhKSsKQIUPQpUsX9jhI+UNCAjB0aOr4kiUAr7QSERERFQg6TbTOnj2Lxo0bS69HjRoFAPDx8cHUqVOxc+dOAECVKlVk8x06dAiNGjUCAKxfvx5DhgxBkyZNoKenhw4dOuDHH3/8KPETfTCtFvj33//GiYiIiKhA0Gmi1ahRIwghMp2e1bQ0VlZW2LBhg5JhERERERERfZA8/YwWERERERFRfsREi4iIiIiISGFMtIiIiIiIiBTGRIuIiIiIiEhhef4Pi4kKPCMjXUdARERERApjokWkS8bGwPHjuo6CiIiIiBTGWweJiIiIiIgUxkSLiIiIiIhIYUy0iHQpMREYPjx1SEzUdTREREREpBA+o0WkSykpwIkT/40TERERUYHAK1pEREREREQKY6JFRERERESkMCZaRERERERECmOiRUREREREpDAmWkRERERERApjokVERERERKQwdu9OpEvGxsDZs7qOgoiIiIgUxitaRERERERECmOiRUREREREpDAmWkS6lJgIjB+fOiQm6joaIiIiIlIIEy0iXUpJAQIDU4eUFF1HQ0REREQKYaJFRERERESkMCZaRERERERECmOiRUREREREpDAmWkRERERERApjokVERERERKQwJlpEREREREQKM9B1AESfNCMj4Nix/8aJiIiIqEBgokWkSyoVYGys6yiIiIiISGG8dZCIiIiIiEhhvKJFpEuJicDMmanj334LqNW6jYeIiIiIFMErWkS6lJIC7N6dOqSk6DoaIiIiIlIIEy0iIiIiIiKFMdEiIiIiIiJSGBMtIiIiIiIihTHRIiIiIiIiUhgTLSIiIiIiIoXpNNE6evQoWrduDUdHR6hUKuzYsUM2XQiByZMnw8HBAcbGxvD09ERISIisTmRkJLp37w5zc3NYWlqib9++iI2N/YhrQUREREREJKfTROvVq1eoXLkyli1bluH0OXPm4Mcff4Sfnx+CgoJgYmICLy8vxMfHS3W6d++Oq1evYv/+/di9ezeOHj2K/v37f6xVIPowRkbA/v2pg5GRrqMhIiIiIoWohBBC10EAgEqlwvbt29G2bVsAqVezHB0dMXr0aIwZMwYAEB0dDTs7O6xZswZdunTB9evXUa5cOZw5cwY1atQAAAQEBKBly5Z4+PAhHB0ds7XsmJgYWFhYIDo6Gubm5rmyfjnhPGFPrrV9d1arXGubiIiIiCi/Uyo3yLPPaIWGhiIsLAyenp5SmYWFBdzd3XHy5EkAwMmTJ2FpaSklWQDg6ekJPT09BAUFZdp2QkICYmJiZAMREREREZFS8myiFRYWBgCws7OTldvZ2UnTwsLCYGtrK5tuYGAAKysrqU5GfH19YWFhIQ1OTk4KR0+UTYmJwOzZqUNioq6jISIiIiKF5NlEKzdNnDgR0dHR0vDgwQNdh0SfqpQUYMuW1CElRdfREBEREZFC8myiZW9vDwAIDw+XlYeHh0vT7O3tERERIZuenJyMyMhIqU5GNBoNzM3NZQMREREREZFS8myi5eLiAnt7ewQGBkplMTExCAoKgoeHBwDAw8MDUVFROHfunFTn4MGD0Gq1cHd3/+gxExERERERAYCBLhceGxuLW7duSa9DQ0Nx4cIFWFlZoXjx4hgxYgSmT58ONzc3uLi4YNKkSXB0dJR6JixbtiyaN2+Ofv36wc/PD0lJSRgyZAi6dOmS7R4HiYiIiIiIlKbTROvs2bNo3Lix9HrUqFEAAB8fH6xZswbjxo3Dq1ev0L9/f0RFRaFevXoICAiA0Rv/N7R+/XoMGTIETZo0gZ6eHjp06IAff/zxo68LERERERFRmjzzP1q6xP/RIp2JiwPq108dP3YMMDbWbTxEREREn7gC/z9aRERERERE+ZVObx0k+uRpNMDOnf+NExEREVGBwESLSJf09AB23EJERERU4PDWQSIiIiIiIoXxihaRLiUlAcuXp44PGgQYGuo2HiIiIiJSBK9oEelScjLw66+pQ3KyrqMhIiIiIoUw0SIiIiIiIlIYEy0iIiIiIiKF5TjRunPnTm7EQUREREREVGDkONEqVaoUGjdujN9++w3x8fG5ERMREREREVG+luNE699//0WlSpUwatQo2NvbY8CAATh9+nRuxEZERERERJQv5TjRqlKlChYvXozHjx9j9erVePLkCerVq4cKFSpgwYIFePr0aW7ESURERERElG+8d2cYBgYGaN++PbZs2YLZs2fj1q1bGDNmDJycnNCzZ088efJEyTiJCiaNBti8OXXQaHQdDREREREp5L0TrbNnz2LQoEFwcHDAggULMGbMGNy+fRv79+/H48eP0aZNGyXjJCqY9PSAkiVTBz12AkpERERUUBjkdIYFCxbA398fwcHBaNmyJdatW4eWLVtC7/+/JLq4uGDNmjVwdnZWOlYiIiIiIqJ8IceJ1ooVK9CnTx/06tULDg4OGdaxtbXFqlWrPjg4ogIvKQnw908d790bMDTUbTxEREREpIgcJ1ohISHvrKNWq+Hj4/NeARF9UpKTgZUrU8d79GCiRURERFRA5PihEH9/f2zZsiVd+ZYtW7B27VpFgiIiIiIiIsrPcpxo+fr6okiRIunKbW1tMXPmTEWCIiIiIiIiys9ynGjdv38fLi4u6cpLlCiB+/fvKxIUERERERFRfpbjRMvW1haXLl1KV37x4kVYW1srEhQREREREVF+luNEq2vXrhg2bBgOHTqElJQUpKSk4ODBgxg+fDi6dOmSGzESERERERHlKznudfCHH37A3bt30aRJExgYpM6u1WrRs2dPPqNFRERERESE90i01Go1Nm3ahB9++AEXL16EsbExKlasiBIlSuRGfEQFm0YDrFv33zgRERERFQg5TrTSfPbZZ/jss8+UjIXo06OnB5Qrp+soiIiIiEhhOU60UlJSsGbNGgQGBiIiIgJarVY2/eDBg4oFR0RERERElB/lONEaPnw41qxZg1atWqFChQpQqVS5ERfRpyEpCfj999Txrl0BQ0PdxkNEREREishxorVx40Zs3rwZLVu2zI14iD4tycnAjz+mjnfqxESLiIiIqIDIcffuarUapUqVyo1YiIiIiIiICoQcJ1qjR4/G4sWLIYTIjXiIiIiIiIjyvRzfOnj8+HEcOnQI+/btQ/ny5WH41q1O27ZtUyw4IiIiIiKi/CjHiZalpSXatWuXG7EQEREREREVCDlOtPz9/XMjDiIiIiIiogIjx89oAUBycjIOHDiAn376CS9fvgQAPH78GLGxsYoGR0RERERElB/l+IrWvXv30Lx5c9y/fx8JCQlo2rQpzMzMMHv2bCQkJMDPzy834iQqmDQa4Kef/hsnIiIiogIhx1e0hg8fjho1auDFixcwNjaWytu1a4fAwEBFgyMq8PT0gOrVUwe997rATERERER5UI6vaB07dgz//PMP1Gq1rNzZ2RmPHj1SLDAiIiIiIqL8Ksc/oWu1WqSkpKQrf/jwIczMzBQJKk1KSgomTZoEFxcXGBsbw9XVFT/88IPsP7yEEJg8eTIcHBxgbGwMT09PhISEKBoHUa5JTgY2b04dkpN1HQ0RERERKSTHiVazZs2waNEi6bVKpUJsbCymTJmCli1bKhkbZs+ejRUrVmDp0qW4fv06Zs+ejTlz5mDJkiVSnTlz5uDHH3+En58fgoKCYGJiAi8vL8THxysaC1GuSEoC5sxJHZKSdB0NERERESkkx7cOzp8/H15eXihXrhzi4+PRrVs3hISEoEiRIvj9998VDe6ff/5BmzZt0KpVKwCptyf+/vvvOH36NIDUq1mLFi3C//73P7Rp0wYAsG7dOtjZ2WHHjh3o0qWLovEQERERERFlR46vaBUrVgwXL17Et99+i5EjR6Jq1aqYNWsWzp8/D1tbW0WDq1OnDgIDA3Hz5k0AwMWLF3H8+HG0aNECABAaGoqwsDB4enpK81hYWMDd3R0nT57MtN2EhATExMTIBiIiIiIiIqXk+IoWABgYGOCrr75SOpZ0JkyYgJiYGJQpUwb6+vpISUnBjBkz0L17dwBAWFgYAMDOzk42n52dnTQtI76+vpg2bVruBU5ERERERJ+0HCda69aty3J6z5493zuYt23evBnr16/Hhg0bUL58eVy4cAEjRoyAo6MjfHx83rvdiRMnYtSoUdLrmJgYODk5KREyERERERFRzhOt4cOHy14nJSXh9evXUKvVKFSokKKJ1tixYzFhwgTpWauKFSvi3r178PX1hY+PD+zt7QEA4eHhcHBwkOYLDw9HlSpVMm1Xo9FAwz+HJSIiIiKiXJLjZ7RevHghG2JjYxEcHIx69eop3hnG69evoffWn7jq6+tDq9UCAFxcXGBvby/7o+SYmBgEBQXBw8ND0ViIiIiIiIiy672e0Xqbm5sbZs2aha+++go3btxQokkAQOvWrTFjxgwUL14c5cuXx/nz57FgwQL06dMHQGrX8iNGjMD06dPh5uYGFxcXTJo0CY6Ojmjbtq1icRDlGrUaSPu7hLf+BJyIiIiI8i9FEi0gtYOMx48fK9UcAGDJkiWYNGkSBg0ahIiICDg6OmLAgAGYPHmyVGfcuHF49eoV+vfvj6ioKNSrVw8BAQEwMjJSNBaiXKGvD9Srp+soiIiIiEhhKiGEyMkMO3fulL0WQuDJkydYunQpnJycsG/fPkUD/BhiYmJgYWGB6OhomJub6zocOE/Yk2tt353VKtfaJiIiIiLK75TKDXJ8RevtW/JUKhVsbGzw+eefY/78+e8dCNEnKTkZSPtxokULwECxi8xEREREpEM5/laX1hEFESkgKQlI+083T08mWkREREQFRI57HSQiIiIiIqKs5fjn8zf/6PddFixYkNPmiYiIiIiI8r0cJ1rnz5/H+fPnkZSUhNKlSwMAbt68CX19fVSrVk2qp1KplIuSiIiIiIgoH8lxotW6dWuYmZlh7dq1KFy4MIDUPzHu3bs36tevj9GjRyseJBERERERUX6S42e05s+fD19fXynJAoDChQtj+vTp7HWQiIiIiIgI75FoxcTE4OnTp+nKnz59ipcvXyoSFBERERERUX6W41sH27Vrh969e2P+/PmoVasWACAoKAhjx45F+/btFQ+QqEBTq4FZs/4bJyIiIqICIceJlp+fH8aMGYNu3bohKSkptREDA/Tt2xdz585VPECiAk1fP/X/s4iIiIioQMlxolWoUCEsX74cc+fOxe3btwEArq6uMDExUTw4IiIiIiKi/Oi9/7D4yZMnePLkCdzc3GBiYgIhhJJxEX0aUlKAAwdSh5QUXUdDRERERArJ8RWt58+fo3Pnzjh06BBUKhVCQkJQsmRJ9O3bF4ULF2bPg0Q5kZgITJiQOn7sGGBsrNt4iIiIiEgROb6iNXLkSBgaGuL+/fsoVKiQVO7t7Y2AgABFgyMiIiIiIsqPcnxF6++//8Zff/2FYsWKycrd3Nxw7949xQIjIiIiIiLKr3J8RevVq1eyK1lpIiMjodFoFAmKiIiIiIgoP8txolW/fn2sW7dOeq1SqaDVajFnzhw0btxY0eCIiIiIiIjyoxzfOjhnzhw0adIEZ8+eRWJiIsaNG4erV68iMjISJ06cyI0YiYiIiIiI8pUcX9GqUKECbt68iXr16qFNmzZ49eoV2rdvj/Pnz8PV1TU3YiQiIiIiIspXcnRFKykpCc2bN4efnx++++673IqJcpHzhD252v7dWa1ytf0Cx9AQmDLlv3EiIiIiKhBylGgZGhri0qVLuRUL0afHwABo3VrXURARERGRwnJ86+BXX32FVatW5UYsREREREREBUKOO8NITk7G6tWrceDAAVSvXh0mJiay6QsWLFAsOKICLyUFOHkyddzDA9DX1208RERERKSIbCValy5dQoUKFaCnp4crV66gWrVqAICbN2/K6qlUKuUjJCrIEhOBESNSx48dA4yNdRoOERERESkjW4lW1apV8eTJE9ja2uLevXs4c+YMrK2tczs2IiIiIiKifClbz2hZWloiNDQUAHD37l1otdpcDYqIiIiIiCg/y9YVrQ4dOqBhw4ZwcHCASqVCjRo1oJ/JsyR37txRNEAiIiIiIqL8JluJ1sqVK9G+fXvcunULw4YNQ79+/WBmZpbbsREREREREeVL2e51sHnz5gCAc+fOYfjw4Uy0iIiIiIiIMpHj7t39/f1zIw4iIiIiIqICI8eJFhEpyNAQGDfuv3EiIiIiKhCYaBHpkoEB0LmzrqMgIiIiIoVlq3t3IiIiIiIiyj5e0SLSJa0WOH8+dbxqVUCPv30QERERFQRMtIh0KSEBGDAgdfzYMcDYWLfxEBEREZEi+PM5ERERERGRwphoERERERERKSzPJ1qPHj3CV199BWtraxgbG6NixYo4e/asNF0IgcmTJ8PBwQHGxsbw9PRESEiIDiMmIiIiIqJPXZ5OtF68eIG6devC0NAQ+/btw7Vr1zB//nwULlxYqjNnzhz8+OOP8PPzQ1BQEExMTODl5YX4+HgdRk5ERERERJ+yPN0ZxuzZs+Hk5AR/f3+pzMXFRRoXQmDRokX43//+hzZt2gAA1q1bBzs7O+zYsQNdunT56DETERERERHl6StaO3fuRI0aNdCpUyfY2tqiatWq+Pnnn6XpoaGhCAsLg6enp1RmYWEBd3d3nDx5MtN2ExISEBMTIxuIiIiIiIiUkqcTrTt37mDFihVwc3PDX3/9hYEDB2LYsGFYu3YtACAsLAwAYGdnJ5vPzs5OmpYRX19fWFhYSIOTk1PurQRRVgwMgGHDUgeDPH2BmYiIiIhyIE9/s9NqtahRowZmzpwJAKhatSquXLkCPz8/+Pj4vHe7EydOxKhRo6TXMTExTLZINwwNgZ49dR0FERERESksT1/RcnBwQLly5WRlZcuWxf379wEA9vb2AIDw8HBZnfDwcGlaRjQaDczNzWUDERERERGRUvJ0olW3bl0EBwfLym7evIkSJUoASO0Yw97eHoGBgdL0mJgYBAUFwcPD46PGSvRetFrg2rXUQavVdTREREREpJA8fevgyJEjUadOHcycOROdO3fG6dOnsXLlSqxcuRIAoFKpMGLECEyfPh1ubm5wcXHBpEmT4OjoiLZt2+o2eKLsSEj479bBY8cAY2PdxkNEREREisjTiVbNmjWxfft2TJw4Ed9//z1cXFywaNEidO/eXaozbtw4vHr1Cv3790dUVBTq1auHgIAAGBkZ6TByIiIiIiL6lKmEEELXQehaTEwMLCwsEB0dnSee13KesEfXIby3u7Na6TqE/CUuDqhfP3WcV7SIiIiIdE6p3CBPP6NFRERERESUHzHRIiIiIiIiUhgTLSIiIiIiIoUx0SIiIiIiIlJYnu51kKjAMzAA+vf/b5yIiIiICgR+syPSJUPD/xItIiIiIioweOsgERERERGRwnhFi0iXtFrg7t3UcWdnQI+/fRAREREVBEy0iHQpIQHo3Dl1nH9YTERERFRg8OdzIiIiIiIihTHRIiIiIiIiUhgTLSIiIiIiIoUx0SIiIiIiIlIYEy0iIiIiIiKFMdEiIiIiIiJSGLt3J9IlAwOgR4//xomIiIioQOA3OyJdMjQEhg/XdRREREREpDDeOkhERERERKQwXtEi0iWtFggLSx23twf0+NsHERERUUHARItIlxISgC+/TB0/dgwwNtZtPERERESkCP58TkREREREpDAmWkRERERERApjokVERERERKQwJlpEREREREQKY6JFRERERESkMCZaRERERERECmP37kS6pK8PdOr03zgRERERFQhMtIh0Sa0Gxo/XdRREREREpDDeOkhERERERKQwXtEi0iUhgKio1HFLS0Cl0mU0RERERKQQJlpEuhQfDzRtmjp+7BhgbKzbeIiIiIhIEbx1kIiIiIiISGFMtIiIiIiIiBTGRIuIiIiIiEhhTLSIiIiIiIgUxkSLiIiIiIhIYUy0iIiIiIiIFJavEq1Zs2ZBpVJhxIgRUll8fDwGDx4Ma2trmJqaokOHDggPD9ddkEQ5oa8PfPFF6qCvr+toiIiIiEgh+SbROnPmDH766SdUqlRJVj5y5Ejs2rULW7ZswZEjR/D48WO0b99eR1ES5ZBaDUydmjqo1bqOhoiIiIgUki8SrdjYWHTv3h0///wzChcuLJVHR0dj1apVWLBgAT7//HNUr14d/v7++Oeff3Dq1CkdRkxERERERJ+yfJFoDR48GK1atYKnp6es/Ny5c0hKSpKVlylTBsWLF8fJkyczbS8hIQExMTGygUgnhADi4lIHIXQdDREREREpxEDXAbzLxo0b8e+//+LMmTPppoWFhUGtVsPS0lJWbmdnh7CwsEzb9PX1xbRp05QOlSjn4uOB+vVTx48dA4yNdRsPERERESkiT1/RevDgAYYPH47169fDyMhIsXYnTpyI6OhoaXjw4IFibRMREREREeXpROvcuXOIiIhAtWrVYGBgAAMDAxw5cgQ//vgjDAwMYGdnh8TERERFRcnmCw8Ph729fabtajQamJubywYiIiIiIiKl5OlbB5s0aYLLly/Lynr37o0yZcpg/PjxcHJygqGhIQIDA9GhQwcAQHBwMO7fvw8PDw9dhExERERERJS3Ey0zMzNUqFBBVmZiYgJra2upvG/fvhg1ahSsrKxgbm6OoUOHwsPDA7Vr19ZFyERERERERHk70cqOhQsXQk9PDx06dEBCQgK8vLywfPlyXYdFRERERESfsHyXaB0+fFj22sjICMuWLcOyZct0ExAREREREdFb8l2iRVSg6OsDTZr8N05EREREBQITLSJdUquB2bN1HQURERERKSxPd+9ORERERESUHzHRIiIiIiIiUhgTLSJdiosDatRIHeLidB0NERERESmEiRYREREREZHCmGgREREREREpjIkWERERERGRwphoERERERERKYyJFhERERERkcKYaBERERERESnMQNcBEH3S9PWBunX/GyciIiKiAoGJFpEuqdXA4sW6joKIiIiIFMZbB4mIiIiIiBTGRIuIiIiIiEhhTLSIdCkuDqhXL3WIi9N1NERERESkED6jRaRr8fG6joCIiIiIFMYrWkRERERERApjokVERERERKQwJlpEREREREQKY6JFRERERESkMCZaRERERERECmOvg6Qo5wl7cq3tu7Na5VrbOqOnB1Sr9t84ERERERUITLSIdEmjAVau1HUURERERKQw/oRORERERESkMCZaRERERERECmOiRaRLcXGAp2fqEBen62iIiIiISCF8RotI16KidB0BERERESmMV7SIiIiIiIgUxkSLiIiIiIhIYUy0iIiIiIiIFMZEi4iIiIiISGFMtIiIiIiIiBTGXgeJdElPDyhX7r9xIiIiIioQmGgR6ZJGA6xbp+soiIiIiEhh/AmdiIiIiIhIYUy0iIiIiIiIFJbnEy1fX1/UrFkTZmZmsLW1Rdu2bREcHCyrEx8fj8GDB8Pa2hqmpqbo0KEDwsPDdRQxUQ7ExwOtW6cO8fG6joaIiIiIFJLnE60jR45g8ODBOHXqFPbv34+kpCQ0a9YMr169kuqMHDkSu3btwpYtW3DkyBE8fvwY7du312HURNkkBPDkSeoghK6jISIiIiKF5PnOMAICAmSv16xZA1tbW5w7dw4NGjRAdHQ0Vq1ahQ0bNuDzzz8HAPj7+6Ns2bI4deoUateurYuwiYiIiIjoE5bnr2i9LTo6GgBgZWUFADh37hySkpLg6ekp1SlTpgyKFy+OkydPZthGQkICYmJiZAMREREREZFS8lWipdVqMWLECNStWxcVKlQAAISFhUGtVsPS0lJW187ODmFhYRm24+vrCwsLC2lwcnLK7dCJiIiIiOgTkq8SrcGDB+PKlSvYuHHjB7UzceJEREdHS8ODBw8UipCIiIiIiCgfPKOVZsiQIdi9ezeOHj2KYsWKSeX29vZITExEVFSU7KpWeHg47O3tM2xLo9FAo9HkdshERERERPSJyvNXtIQQGDJkCLZv346DBw/CxcVFNr169eowNDREYGCgVBYcHIz79+/Dw8PjY4dLlDMqFVCyZOqgUuk6GiIiIiJSSJ6/ojV48GBs2LABf/75J8zMzKTnriwsLGBsbAwLCwv07dsXo0aNgpWVFczNzTF06FB4eHiwx0HK+4yMgM2bdR0FERERESkszydaK1asAAA0atRIVu7v749evXoBABYuXAg9PT106NABCQkJ8PLywvLlyz9ypERERERERKnyfKIlsvEnrkZGRli2bBmWLVv2ESIiIiIiIiLKWp5/RouoQIuPBzp3Th3i43UdDREREREpJM9f0SIq0IQA7tz5b5yIiIiICgRe0SIiIiIiIlIYEy0iIiIiIiKFMdEiIiIiIiJSGBMtIiIiIiIihTHRIiIiIiIiUhh7HSTSJZUKcHD4b5yIiIiICgQmWkS6ZGQE7Nql6yiIiIiISGG8dZCIiIiIiEhhTLSIiIiIiIgUxkSLSJcSEoCePVOHhARdR0NERERECuEzWkQfgfOEPRmWa5ISsGX/SQBAp2/3IMFQ817t353V6r1jIyIiIiLl8YoWERERERGRwphoERERERERKYyJFhERERERkcKYaBERERERESmMnWFQvpFZhxJKYYcSRERERKQUJlpEOhajMdF1CERERESkMCZaRDqUYKjBV11m6DoMIiIiIlIYn9EiIiIiIiJSGBMtIiIiIiIihfHWQSIdUicnYeoBPwDAVM9vkGhgqOOIiIiIiEgJTLSIdEgltKgQflsaJyIiIqKCgYkW0f/L7e7jiYiIiOjTwWe0iIiIiIiIFMZEi4iIiIiISGFMtIiIiIiIiBTGRIuIiIiIiEhh7AyDSMcS9NW6DoGIiIiIFMZEi0iHEgw16PTVHF2HQUREREQK462DRERERERECmOiRUREREREpDDeOkikQ4YpSfj20GoAwMzGfZCkb6jjiIiIiIhICUy0iHRIT6tF9UfXpXHov187zhP2KBiV3N1ZrXKtbSIiIqKCircOEhERERERKYyJFhERERERkcIKTKK1bNkyODs7w8jICO7u7jh9+rSuQyIiIiIiok9UgXhGa9OmTRg1ahT8/Pzg7u6ORYsWwcvLC8HBwbC1tdV1eEREn5TcfGYQ4HODRES6wvN7zhSIK1oLFixAv3790Lt3b5QrVw5+fn4oVKgQVq9erevQiIiIiIjoE5Tvr2glJibi3LlzmDhxolSmp6cHT09PnDx5MsN5EhISkJCQIL2Ojo4GAMTExORusNmkTXit6xDoI0lJSkCsVps6nvAaWm2KjiNKL6+8Lyj/yO1zGI9JIiLd+FTO72lxCCE+qJ18n2g9e/YMKSkpsLOzk5Xb2dnhxo0bGc7j6+uLadOmpSt3cnLKlRiJslI3bWRZD12GkSmLRbqOgEiOxyQRUcGU187vL1++hIWFxXvPn+8TrfcxceJEjBo1Snqt1WoRGRkJa2trqFQqHUaWmkE7OTnhwYMHMDc312kslB73T97G/ZP3cR/lbdw/eRv3T97G/ZP3ZXcfCSHw8uVLODo6ftDy8n2iVaRIEejr6yM8PFxWHh4eDnt7+wzn0Wg00Gg0sjJLS8vcCvG9mJub802ah3H/5G3cP3kf91Hexv2Tt3H/5G3cP3lfdvbRh1zJSpPvO8NQq9WoXr06AgMDpTKtVovAwEB4eHjoMDIiIiIiIvpU5fsrWgAwatQo+Pj4oEaNGqhVqxYWLVqEV69eoXfv3roOjYiIiIiIPkEFItHy9vbG06dPMXnyZISFhaFKlSoICAhI10FGfqDRaDBlypR0tzZS3sD9k7dx/+R93Ed5G/dP3sb9k7dx/+R9H3sfqcSH9ltIREREREREMvn+GS0iIiIiIqK8hokWERERERGRwphoERERERERKYyJFhERERERkcKYaOUhy5Ytg7OzM4yMjODu7o7Tp0/rOqRPwtSpU6FSqWRDmTJlpOnx8fEYPHgwrK2tYWpqig4dOqT7g+z79++jVatWKFSoEGxtbTF27FgkJyd/7FUpEI4ePYrWrVvD0dERKpUKO3bskE0XQmDy5MlwcHCAsbExPD09ERISIqsTGRmJ7t27w9zcHJaWlujbty9iY2NldS5duoT69evDyMgITk5OmDNnTm6vWoHxrn3Uq1evdO+p5s2by+pwH+UOX19f1KxZE2ZmZrC1tUXbtm0RHBwsq6PUOe3w4cOoVq0aNBoNSpUqhTVr1uT26hUI2dlHjRo1Svce+uabb2R1uI9yx4oVK1CpUiXpD209PDywb98+aTrfP7r1rv2T5947gvKEjRs3CrVaLVavXi2uXr0q+vXrJywtLUV4eLiuQyvwpkyZIsqXLy+ePHkiDU+fPpWmf/PNN8LJyUkEBgaKs2fPitq1a4s6depI05OTk0WFChWEp6enOH/+vNi7d68oUqSImDhxoi5WJ9/bu3ev+O6778S2bdsEALF9+3bZ9FmzZgkLCwuxY8cOcfHiRfHll18KFxcXERcXJ9Vp3ry5qFy5sjh16pQ4duyYKFWqlOjatas0PTo6WtjZ2Ynu3buLK1euiN9//10YGxuLn3766WOtZr72rn3k4+MjmjdvLntPRUZGyupwH+UOLy8v4e/vL65cuSIuXLggWrZsKYoXLy5iY2OlOkqc0+7cuSMKFSokRo0aJa5duyaWLFki9PX1RUBAwEdd3/woO/uoYcOGol+/frL3UHR0tDSd+yj37Ny5U+zZs0fcvHlTBAcHi2+//VYYGhqKK1euCCH4/tG1d+2fvPbeYaKVR9SqVUsMHjxYep2SkiIcHR2Fr6+vDqP6NEyZMkVUrlw5w2lRUVHC0NBQbNmyRSq7fv26ACBOnjwphEj90qmnpyfCwsKkOitWrBDm5uYiISEhV2Mv6N7+Eq/VaoW9vb2YO3euVBYVFSU0Go34/fffhRBCXLt2TQAQZ86ckers27dPqFQq8ejRIyGEEMuXLxeFCxeW7Z/x48eL0qVL5/IaFTyZJVpt2rTJdB7uo48nIiJCABBHjhwRQih3Ths3bpwoX768bFne3t7Cy8srt1epwHl7HwmR+mVx+PDhmc7DffRxFS5cWPzyyy98/+RRaftHiLz33uGtg3lAYmIizp07B09PT6lMT08Pnp6eOHnypA4j+3SEhITA0dERJUuWRPfu3XH//n0AwLlz55CUlCTbN2XKlEHx4sWlfXPy5ElUrFhR9gfZXl5eiImJwdWrVz/uihRwoaGhCAsLk+0PCwsLuLu7y/aHpaUlatSoIdXx9PSEnp4egoKCpDoNGjSAWq2W6nh5eSE4OBgvXrz4SGtTsB0+fBi2trYoXbo0Bg4ciOfPn0vTuI8+nujoaACAlZUVAOXOaSdPnpS1kVaHn1k59/Y+SrN+/XoUKVIEFSpUwMSJE/H69WtpGvfRx5GSkoKNGzfi1atX8PDw4Psnj3l7/6TJS+8dgxzPQYp79uwZUlJSZDsdAOzs7HDjxg0dRfXpcHd3x5o1a1C6dGk8efIE06ZNQ/369XHlyhWEhYVBrVbD0tJSNo+dnR3CwsIAAGFhYRnuu7RppJy07ZnR9n5zf9ja2sqmGxgYwMrKSlbHxcUlXRtp0woXLpwr8X8qmjdvjvbt28PFxQW3b9/Gt99+ixYtWuDkyZPQ19fnPvpItFotRowYgbp166JChQoAoNg5LbM6MTExiIuLg7GxcW6sUoGT0T4CgG7duqFEiRJwdHTEpUuXMH78eAQHB2Pbtm0AuI9y2+XLl+Hh4YH4+HiYmppi+/btKFeuHC5cuMD3Tx6Q2f4B8t57h4kWffJatGghjVeqVAnu7u4oUaIENm/ezJMd0Xvo0qWLNF6xYkVUqlQJrq6uOHz4MJo0aaLDyD4tgwcPxpUrV3D8+HFdh0KZyGwf9e/fXxqvWLEiHBwc0KRJE9y+fRuurq4fO8xPTunSpXHhwgVER0dj69at8PHxwZEjR3QdFv2/zPZPuXLl8tx7h7cO5gFFihSBvr5+ul5rwsPDYW9vr6OoPl2Wlpb47LPPcOvWLdjb2yMxMRFRUVGyOm/uG3t7+wz3Xdo0Uk7a9szqvWJvb4+IiAjZ9OTkZERGRnKf6UjJkiVRpEgR3Lp1CwD30ccwZMgQ7N69G4cOHUKxYsWkcqXOaZnVMTc35w9U2ZTZPsqIu7s7AMjeQ9xHuUetVqNUqVKoXr06fH19UblyZSxevJjvnzwis/2TEV2/d5ho5QFqtRrVq1dHYGCgVKbVahEYGCi755Q+jtjYWNy+fRsODg6oXr06DA0NZfsmODgY9+/fl/aNh4cHLl++LPviuH//fpibm0uXskkZLi4usLe3l+2PmJgYBAUFyfZHVFQUzp07J9U5ePAgtFqtdML18PDA0aNHkZSUJNXZv38/SpcuzVvScsHDhw/x/PlzODg4AOA+yk1CCAwZMgTbt2/HwYMH091+qdQ5zcPDQ9ZGWh1+Zr3bu/ZRRi5cuAAAsvcQ99HHo9VqkZCQwPdPHpW2fzKi8/dOjrvPoFyxceNGodFoxJo1a8S1a9dE//79haWlpaxXFModo0ePFocPHxahoaHixIkTwtPTUxQpUkREREQIIVK7ci1evLg4ePCgOHv2rPDw8BAeHh7S/GldhTZr1kxcuHBBBAQECBsbG3bv/p5evnwpzp8/L86fPy8AiAULFojz58+Le/fuCSFSu3e3tLQUf/75p7h06ZJo06ZNht27V61aVQQFBYnjx48LNzc3WdfhUVFRws7OTvTo0UNcuXJFbNy4URQqVIhdh2dTVvvo5cuXYsyYMeLkyZMiNDRUHDhwQFSrVk24ubmJ+Ph4qQ3uo9wxcOBAYWFhIQ4fPizr3vj169dSHSXOaWndH48dO1Zcv35dLFu2jN1TZ9O79tGtW7fE999/L86ePStCQ0PFn3/+KUqWLCkaNGggtcF9lHsmTJggjhw5IkJDQ8WlS5fEhAkThEqlEn///bcQgu8fXctq/+TF9w4TrTxkyZIlonjx4kKtVotatWqJU6dO6TqkT4K3t7dwcHAQarVaFC1aVHh7e4tbt25J0+Pi4sSgQYNE4cKFRaFChUS7du3EkydPZG3cvXtXtGjRQhgbG4siRYqI0aNHi6SkpI+9KgXCoUOHBIB0g4+PjxAitYv3SZMmCTs7O6HRaESTJk1EcHCwrI3nz5+Lrl27ClNTU2Fubi569+4tXr58Katz8eJFUa9ePaHRaETRokXFrFmzPtYq5ntZ7aPXr1+LZs2aCRsbG2FoaChKlCgh+vXrl+5HI+6j3JHRfgEg/P39pTpKndMOHTokqlSpItRqtShZsqRsGZS5d+2j+/fviwYNGggrKyuh0WhEqVKlxNixY2X/BSQE91Fu6dOnjyhRooRQq9XCxsZGNGnSREqyhOD7R9ey2j958b2jEkKInF8HIyIiIiIioszwGS0iIiIiIiKFMdEiIiIiIiJSGBMtIiIiIiIihTHRIiIiIiIiUhgTLSIiIiIiIoUx0SIiIiIiIlIYEy0iIiIiIiKFMdEiIiIiIiJSGBMtIiLKdWvWrIGlpWWm0+/evQuVSoULFy58tJgKsndt748pMDAQZcuWRUpKiq5DydSzZ89ga2uLhw8f6joUIipAmGgREb3h5MmT0NfXR6tWrXQdClGBMG7cOPzvf/+Dvr6+rkPJVJEiRdCzZ09MmTJF16EQUQHCRIuI6A2rVq3C0KFDcfToUTx+/DjLukIIJCcnf6TIdC8lJQVarVbXYZAOvO+xfvz4cdy+fRsdOnTIhahSKfU+7N27N9avX4/IyEgFoiIiYqJFRCSJjY3Fpk2bMHDgQLRq1Qpr1qyRTT98+DBUKhX27duH6tWrQ6PR4Pjx49BqtfD19YWLiwuMjY1RuXJlbN26VZovJSUFffv2laaXLl0aixcvfmc8V69exRdffAFzc3OYmZmhfv36uH37NgBAq9Xi+++/R7FixaDRaFClShUEBARI89apUwfjx4+Xtff06VMYGhri6NGjAICEhASMGTMGRYsWhYmJCdzd3XH48GGpftrtZzt37kS5cuWg0Whw//79d86XNm/x4sVRqFAhtGvXDs+fP8/OLsCNGzdQp04dGBkZoUKFCjhy5AiA1C/TpUqVwrx582T1L1y4AJVKhVu3bmXY3uHDh1GrVi2YmJjA0tISdevWxb179wAAU6dORZUqVfDrr7/C2dkZFhYW6NKlC16+fCnNn5CQgGHDhsHW1hZGRkaoV68ezpw5I02vUaOGLKa2bdvC0NAQsbGxAICHDx9mGV9aDKtXr0bx4sVhamqKQYMGISUlBXPmzIG9vT1sbW0xY8YM2XwLFixAxYoVYWJiAicnJwwaNEhaZkaePn2KGjVqoF27dkhISHjnMZvZsX7x4kU0btwYZmZmMDc3R/Xq1XH27NlMl7tx40Y0bdoURkZGUtnt27fRpk0b2NnZwdTUFDVr1sSBAwdk8yUkJGD8+PFwcnKCRqNBqVKlsGrVqixje9c6vXjxAt27d4eNjQ2MjY3h5uYGf39/aXr58uXh6OiI7du3Z7o+REQ5IoiISAghxKpVq0SNGjWEEELs2rVLuLq6Cq1WK00/dOiQACAqVaok/v77b3Hr1i3x/PlzMX36dFGmTBkREBAgbt++Lfz9/YVGoxGHDx8WQgiRmJgoJk+eLM6cOSPu3LkjfvvtN1GoUCGxadOmTGN5+PChsLKyEu3btxdnzpwRwcHBYvXq1eLGjRtCCCEWLFggzM3Nxe+//y5u3Lghxo0bJwwNDcXNmzeFEEIsXbpUFC9eXBb/kiVLZGVff/21qFOnjjh69Ki4deuWmDt3rtBoNFIb/v7+wtDQUNSpU0ecOHFC3LhxQ7x69eqd8506dUro6emJ2bNni+DgYLF48WJhaWkpLCwsMl3f0NBQAUAUK1ZMbN26VVy7dk18/fXXwszMTDx79kwIIcSMGTNEuXLlZPMNGzZMNGjQIMM2k5KShIWFhRgzZoy4deuWuHbtmlizZo24d++eEEKIKVOmCFNTU9G+fXtx+fJlcfToUWFvby++/fZbWfuOjo5i79694urVq8LHx0cULlxYPH/+XAghxKhRo0SrVq2EEEJotVphZWUlihQpIvbt2yeEEOK3334TRYsWzXS902Lo2LGjuHr1qti5c6dQq9XCy8tLDB06VNy4cUOsXr1aABCnTp2S5lu4cKE4ePCgCA0NFYGBgaJ06dJi4MCB0nR/f39pe9+/f1+ULl1a+Pj4iOTkZCGEeOcxm9mxXr58efHVV1+J69evi5s3b4rNmzeLCxcuZLp+lSpVErNmzZKVXbhwQfj5+YnLly+Lmzdviv/973/CyMhI2i9CCNG5c2fh5OQktm3bJm7fvi0OHDggNm7cmGVs71qnwYMHiypVqogzZ86I0NBQsX//frFz505ZbN7e3sLHxyfT9SEiygkmWkRE/69OnTpi0aJFQojUL+lFihQRhw4dkqanfcHbsWOHVBYfHy8KFSok/vnnH1lbffv2FV27ds10WYMHDxYdOnTIdPrEiROFi4uLSExMzHC6o6OjmDFjhqysZs2aYtCgQUIIISIiIoSBgYE4evSoNN3Dw0OMHz9eCCHEvXv3hL6+vnj06JGsjSZNmoiJEycKIVK/rAOQfZHOznxdu3YVLVu2lE339vbOVqL15pfypKQkUaxYMTF79mwhhBCPHj0S+vr6IigoSAiRmsAWKVJErFmzJsM2nz9/LgBIX7TfNmXKFFGoUCERExMjlY0dO1a4u7sLIYSIjY0VhoaGYv369dL0xMRE4ejoKObMmSOEEGLnzp3CwsJCJCcniwsXLgh7e3sxfPhwaTt//fXXolu3bpmud0YxeHl5CWdnZ5GSkiKVlS5dWvj6+mbazpYtW4S1tbX0Oi3RunHjhnBychLDhg2TEuzsHLMZHetCCGFmZpbp9s6IhYWFWLdu3TvrlS9fXixZskQIIURwcLAAIPbv359h3fd9H7Zu3Vr07t07yzhGjhwpGjVq9M54iYiyw0A319GIiPKW4OBgnD59WrptyMDAAN7e3li1ahUaNWokq1ujRg1p/NatW3j9+jWaNm0qq5OYmIiqVatKr5ctW4bVq1fj/v37iIuLQ2JiIqpUqZJpPBcuXED9+vVhaGiYblpMTAweP36MunXrysrr1q2LixcvAgBsbGzQrFkzrF+/HvXr10doaChOnjyJn376CQBw+fJlpKSk4LPPPpO1kZCQAGtra+m1Wq1GpUqVpNfZme/69eto166dbLqHh4fs1sbMeHh4SOMGBgaoUaMGrl+/DgBwdHREq1atsHr1atSqVQu7du1CQkICOnXqlGFbVlZW6NWrF7y8vNC0aVN4enqic+fOcHBwkOo4OzvDzMxMeu3g4ICIiAgAqbe4JSUlybazoaEhatWqJcVUv359vHz5EufPn8c///yDhg0bolGjRpg1axYA4MiRIxg7dmyW6/x2DHZ2dtDX14eenp6sLC0uADhw4AB8fX1x48YNxMTEIDk5GfHx8Xj9+jUKFSoEAIiLi0P9+vXRrVs3LFq0SJo3u8csID/WAWDUqFH4+uuv8euvv8LT0xOdOnWCq6trpusWFxcnu20QSL1Fd+rUqdizZw+ePHmC5ORkxMXF4f79+wBSj319fX00bNgwq82W4/fhwIED0aFDB/z7779o1qwZ2rZtizp16sjqGxsb4/Xr11kul4gou5hoEREhtROM5ORkODo6SmVCCGg0GixduhQWFhZSuYmJiTSe9lzMnj17ULRoUVmbGo0GQOpzKmPGjMH8+fPh4eEBMzMzzJ07F0FBQZnGY2xs/MHr1L17dwwbNgxLlizBhg0bULFiRVSsWFGKW19fH+fOnUvXG5ypqaksDpVKJb3O7ny55euvv0aPHj2wcOFC+Pv7w9vbW0osMuLv749hw4YhICAAmzZtwv/+9z/s378ftWvXBoB0iaxKpcpRhx+WlpaoXLkyDh8+jJMnT6Jp06Zo0KABvL29cfPmTYSEhLwzYcgohqziunv3Lr744gsMHDgQM2bMgJWVFY4fP46+ffsiMTFR2h4ajQaenp7YvXs3xo4dKx2f2Tlm07x5rAOpz5R169YNe/bswb59+zBlyhRs3LgxXWKdpkiRInjx4oWsbMyYMdi/fz/mzZuHUqVKwdjYGB07dkRiYiKA7B/7OX0ftmjRAvfu3cPevXuxf/9+NGnSBIMHD5Y9YxcZGQkbG5tsLZ+I6F3YGQYRffKSk5Oxbt06zJ8/HxcuXJCGixcvwtHREb///num877ZSUSpUqVkg5OTEwDgxIkTqFOnDgYNGoSqVauiVKlSUqcWmalUqRKOHTuGpKSkdNPMzc3h6OiIEydOyMpPnDiBcuXKSa/btGmD+Ph4BAQEYMOGDejevbs0rWrVqkhJSUFERES6uO3t7TONKzvzlS1bNl0SeerUqSzXN6N6ycnJOHfuHMqWLSuVtWzZEiYmJlixYgUCAgLQp0+fd7ZZtWpVTJw4Ef/88w8qVKiADRs2ZCsWV1dXqNVq2XZOSkrCmTNnZNu5YcOGOHToEI4ePYpGjRrBysoKZcuWxYwZM+Dg4JDu6t+HOnfuHLRaLebPn4/atWvjs88+y7CHTD09Pfz666+oXr06GjduLNXJzjGblc8++wwjR47E33//jfbt28s6lHhb1apVce3aNVnZiRMn0KtXL7Rr1w4VK1aEvb097t69K02vWLEitFqt1BFKdmR3nWxsbODj44PffvsNixYtwsqVK2XtXLlyJd1VPSKi98UrWkT0ydu9ezdevHiBvn37yq5cAUCHDh2watUqfPPNNxnOa2ZmhjFjxmDkyJHQarWoV68eoqOjceLECZibm8PHxwdubm5Yt24d/vrrL7i4uODXX3/FmTNn4OLikmlMQ4YMwZIlS9ClSxdMnDgRFhYWOHXqFGrVqoXSpUtj7NixmDJlClxdXVGlShX4+/vjwoULWL9+vdSGiYkJ2rZti0mTJuH69evo2rWrNO2zzz5D9+7d0bNnT8yfPx9Vq1bF06dPERgYiEqVKmX6P2LZmW/YsGGoW7cu5s2bhzZt2uCvv/7K1m2DQOotlm5ubihbtiwWLlyIFy9eyJIpfX199OrVCxMnToSbm5vsVsO3hYaGYuXKlfjyyy/h6OiI4OBghISEoGfPntmKxcTEBAMHDsTYsWNhZWWF4sWLY86cOXj9+jX69u0r1WvUqBGWLFkCGxsblClTRipbunRpprc1fohSpUohKSkJS5YsQevWrXHixAn4+fllWFdfXx/r169H165d8fnnn+Pw4cOwt7d/5zGbkbi4OIwdOxYdO3aEi4sLHj58iDNnzmTZdbuXlxfWrl0rK3Nzc8O2bdvQunVrqFQqTJo0SXYV0dnZGT4+PujTpw9+/PFHVK5cGffu3UNERAQ6d+6c4XKy8z6cPHkyqlevjvLlyyMhIQG7d++WJfGvX7/GuXPnMHPmzEzXh4goR3T9kBgRka598cUX6TpvSBMUFCQAiIsXL0oP4b948UJWR6vVikWLFonSpUsLQ0NDYWNjI7y8vMSRI0eEEKkP6vfq1UtYWFgIS0tLMXDgQDFhwgRRuXLlLOO6ePGiaNasmShUqJAwMzMT9evXF7dv3xZCCJGSkiKmTp0qihYtKgwNDUXlypWlnu7etHfvXgEgw5750npDdHZ2FoaGhsLBwUG0a9dOXLp0SQgh77kuJ/MJkdqDY7FixYSxsbFo3bq1mDdvXrY6w9iwYYOoVauWUKvVoly5cuLgwYPp6t6+fVsAkDqkyExYWJho27atcHBwEGq1WpQoUUJMnjxZ6mRiypQp6fbBwoULRYkSJaTXcXFxYujQoaJIkSJCo9GIunXritOnT8vmef78uVCpVMLb21sq2759uwAg/Pz8sowxoxh8fHxEmzZtZGUNGzYUw4cPl14vWLBAODg4CGNjY+Hl5SXWrVsnOzbf3ndJSUmiffv2omzZsiI8PPydx2xGx3pCQoLo0qWLcHJyEmq1Wjg6OoohQ4aIuLi4TNfv+fPnwsjISOotU4jUfd24cWNhbGwsnJycxNKlS9OtX1xcnBg5cqS070qVKiVWr16daWxCvPt9+MMPP4iyZcsKY2NjYWVlJdq0aSPu3Lkjzb9hwwZRunTpTNeFiCinVEIIobMsj4iIKIeOHTuGJk2a4MGDB7Czs9N1OPQOY8eORUxMjNQRS15Vu3ZtDBs2DN26ddN1KERUQPAZLSIiyhcSEhLw8OFDTJ06FZ06dWKSlU989913KFGiRI46GfnYnj17hvbt28turyUi+lC8okVERPnCmjVr0LdvX1SpUgU7d+5M17scERFRXsJEi4iIiIiISGG8dZCIiIiIiEhhTLSIiIiIiIgUxkSLiIiIiIhIYUy0iIiIiIiIFMZEi4iIiIiISGFMtIiIiIiIiBTGRIuIiIiIiEhhTLSIiIiIiIgU9n/HxqR21YWq6gAAAABJRU5ErkJggg==\n"},"metadata":{}}],"source":["plot_compare('Snow Making_ac', 'Area covered by snow makers (acres)')"]},{"cell_type":"markdown","metadata":{"id":"mf_Mp_DeXvBd"},"source":["Big Mountain is very high up the league table of snow making area."]},{"cell_type":"markdown","metadata":{"id":"xnDZDOE1XvBd"},"source":["### 5.8.4 Total number of chairs"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"tQZZA50vXvBd","executionInfo":{"status":"ok","timestamp":1721138926936,"user_tz":240,"elapsed":383,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"ccbf2e34-1eec-45b1-bd5d-f4f932b358c9"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAA0kAAAHWCAYAAACi1sL/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABZYklEQVR4nO3dd3gU1f7H8c+m7CaEFAiQgLRIk64GkAhKC4QqTVBEqQr3So9cig1QpMgFAaWoYCiKKFywXkGaiEgvFpSi0q6EYAECgdSd3x/5ZdklCeyGJBuS9+t55nlmzsye+c7s2Um+O2fOmgzDMAQAAAAAkCR5uDsAAAAAAChISJIAAAAAwA5JEgAAAADYIUkCAAAAADskSQAAAABghyQJAAAAAOyQJAEAAACAHZIkAAAAALBDkgQAAAAAdkiSADf56quvZDKZ9NVXX7k7FJdlxL569Wp3h+KUuLg4PfzwwwoODpbJZNLs2bNvqT6TyaShQ4fmTnByf1uoXLmy+vXr55Z4Jk6cKJPJ5FCW2+f3RpYsWSKTyaQTJ07ky/7spaamasyYMapQoYI8PDzUpUuXfI+hqGjevLmaN2/u7jByVUG/Dmd8tvbu3evuUIAcIUlCkWIymZyanPnncMqUKfroo4/yPGbculGjRmn9+vUaP368li9frrZt27o7pEKpIH8mCmJs77zzjmbMmKGHH35YS5cu1ahRo9wdUoHw008/aeLEiW5JXIuy+fPna8mSJe4OAygwvNwdAJCfli9f7rC8bNkybdiwIVN5zZo1b1rXlClT9PDDD/Pt721g8+bN6ty5s0aPHu3uULL04IMP6urVqzKbze4ORVLO48nJZ+L555/XuHHjXIzQddnF9sQTT+jRRx+VxWLJ8xiut3nzZt1xxx167bXX8n3fBdlPP/2kSZMmqXnz5qpcuXKu1Pnll1/mSj2F2fz581WqVCmHu8pAUUaShCLl8ccfd1jeuXOnNmzYkKkcBUNCQoL8/PxuuZ5z584pKCjo1gPKIx4eHvLx8bnpdleuXFGxYsUKTDy3IuO99fLykpeX+/4UeXp6ytPT0y37zu12abValZyc7PR7ZxiGEhMT5evrm2sx3IrExMQ8+6KgoHwBkRtSU1NltVrdHUaBkV/XRRQ9dLcDrpOQkKBnnnlGFSpUkMViUY0aNfTvf/9bhmHYtjGZTEpISNDSpUttXfQyvn07efKknn76adWoUUO+vr4KDg5Wjx49ctx1JOOZjV9++UX9+vVTUFCQAgMD1b9/f125csW23YkTJ2QymbLsLmEymTRx4sRMdR49elSPP/64AgMDVbp0ab3wwgsyDEOnT59W586dFRAQoNDQUM2cOTPL2NLS0vTss88qNDRUfn5+euihh3T69OlM2+3atUtt27ZVYGCgihUrpmbNmmn79u1ZHudPP/2kxx57TCVKlFDTpk1veG5+++039ejRQyVLllSxYsXUuHFjff7557b1GX3iDcPQvHnzbO/VjVitVs2ZM0d169aVj4+PSpcurbZt22bZr/6jjz5SnTp1ZLFYVLt2ba1bt85hvbNtIatngJo3b646depo3759evDBB1WsWDE9++yzkqS9e/cqKipKpUqVkq+vr8LCwjRgwIAbHpeU/k/x5MmTVb58eRUrVkwtWrTQoUOHMm2XVTzHjh1T9+7dFRoaKh8fH5UvX16PPvqoLl68KOnGn4kbvbdZPZOU4b333lONGjXk4+Oj8PBwff311w7r+/Xrl+WdhuvrvFFs2T2TNH/+fNWuXVsWi0XlypXTkCFDdOHCBYdtMt6jn376SS1atFCxYsV0xx136NVXX83yeDJkfFa3bNmiQ4cOZerm68w1KOO4hg4dqvfee88W6/Vt0F7lypXVsWNHrV+/Xg0aNJCvr6/efPNNSdKFCxc0cuRI2z6rVq2q6dOnZ/pnfOXKlQoPD5e/v78CAgJUt25dzZkzx2Gbm30upWttbOXKlXr++ed1xx13qFixYpo7d6569OghSWrRokWmc5PTtn/9M0kZ+//www/1yiuvqHz58vLx8VGrVq30yy+/3LS+W72GJicn68UXX1R4eLgCAwPl5+enBx54QFu2bHHYLqOt/Pvf/9bs2bNVpUoVWSwW/fTTT1nGlZSUpI4dOyowMFDffvutpPRr2uzZs1W7dm35+PgoJCREgwcP1vnz522vq1y5sg4dOqStW7fazvnNnuFypi1kxBQdHa3SpUvLz89PXbt21R9//OGwzccff6wOHTqoXLlyslgsqlKlil5++WWlpaU5bHej62JSUpImTJigqlWrymKxqEKFChozZoySkpJueBxAdriTBNgxDEMPPfSQtmzZooEDB+ruu+/W+vXr9a9//Uu///67rVvM8uXL9eSTT6pRo0YaNGiQJKlKlSqSpD179ujbb7/Vo48+qvLly+vEiRNasGCBmjdvrp9++inH33j17NlTYWFhmjp1qvbv369FixapTJkymj59eo6P95FHHlHNmjU1bdo0ff7555o8ebJKliypN998Uy1bttT06dP13nvvafTo0WrYsKEefPBBh9e/8sorMplMGjt2rM6dO6fZs2crMjJSBw8etH07vXnzZrVr107h4eGaMGGCPDw8FBMTo5YtW2rbtm1q1KiRQ509evRQtWrVNGXKlEz/FNqLi4vT/fffrytXrmj48OEKDg7W0qVL9dBDD2n16tXq2rWrHnzwQS1fvlxPPPGEWrdurT59+tz0nAwcOFBLlixRu3bt9OSTTyo1NVXbtm3Tzp071aBBA9t233zzjdasWaOnn35a/v7+mjt3rrp3765Tp04pODhY0q23hb/++kvt2rXTo48+qscff1whISE6d+6c2rRpo9KlS2vcuHEKCgrSiRMntGbNmpse24svvqjJkyerffv2at++vfbv3682bdooOTn5hq9LTk5WVFSUkpKSNGzYMIWGhur333/XZ599pgsXLigwMPCGn4kMzr63krR161Z98MEHGj58uCwWi+bPn6+2bdtq9+7dqlOnzk2P1Z4zsdmbOHGiJk2apMjISP3zn//UkSNHtGDBAu3Zs0fbt2+Xt7e3bdvz58+rbdu26tatm3r27KnVq1dr7Nixqlu3rtq1a5dl/aVLl9by5cv1yiuv6PLly5o6daqk9G6+zl6DMmzevFkffvihhg4dqlKlSt20e9qRI0fUq1cvDR48WE899ZRq1KihK1euqFmzZvr99981ePBgVaxYUd9++63Gjx+v2NhY20AnGzZsUK9evdSqVSvbdefnn3/W9u3bNWLECEnOfS7tvfzyyzKbzRo9erSSkpLUpk0bDR8+XHPnztWzzz5r6/pcs2bNW2r72Zk2bZo8PDw0evRoXbx4Ua+++qp69+6tXbt2OfX6nF5D4+PjtWjRIvXq1UtPPfWULl26pMWLFysqKkq7d+/W3Xff7bCfmJgYJSYmatCgQbJYLCpZsmSmpP3q1avq3Lmz9u7dq40bN6phw4aSpMGDB2vJkiXq37+/hg8fruPHj+uNN97QgQMHbO159uzZGjZsmIoXL67nnntOkhQSEpLtcTvTFjIMGzZMJUqU0IQJE3TixAnNnj1bQ4cO1QcffGDbZsmSJSpevLiio6NVvHhxbd68WS+++KLi4+M1Y8YMh/qyui5arVY99NBD+uabbzRo0CDVrFlTP/zwg1577TUdPXq0wD2PiNuEARRhQ4YMMew/Bh999JEhyZg8ebLDdg8//LBhMpmMX375xVbm5+dn9O3bN1OdV65cyVS2Y8cOQ5KxbNkyW9mWLVsMScaWLVtuGOOECRMMScaAAQMcyrt27WoEBwfblo8fP25IMmJiYjLVIcmYMGFCpjoHDRpkK0tNTTXKly9vmEwmY9q0abby8+fPG76+vg7HmhH7HXfcYcTHx9vKP/zwQ0OSMWfOHMMwDMNqtRrVqlUzoqKiDKvVatvuypUrRlhYmNG6detMMfXq1euG5yPDyJEjDUnGtm3bbGWXLl0ywsLCjMqVKxtpaWkOxz9kyJCb1rl582ZDkjF8+PBM6+zjl2SYzWaH9vDdd98ZkozXX3/d4Tiv52xbaNasmSHJWLhwocPr165da0gy9uzZc9PjsXfu3DnDbDYbHTp0cDiWZ5991pCU5fubEc+BAwcMScaqVatuuI/sPhM3em8z1tmTZEgy9u7days7efKk4ePjY3Tt2tVW1rdvX6NSpUpO1ZldbDExMYYk4/jx44ZhXDtPbdq0cWhDb7zxhiHJeOedd2xlGe+R/XuZlJRkhIaGGt27d8+0r+s1a9bMqF27tkOZK9cgSYaHh4dx6NChm+7LMAyjUqVKhiRj3bp1DuUvv/yy4efnZxw9etShfNy4cYanp6dx6tQpwzAMY8SIEUZAQICRmpqa7T6c/VxmtLE777wz0+dk1apVWV4bc9r2DSP9XDdr1sy2nLH/mjVrGklJSbbyOXPmGJKMH3744Yb13eo1NDU11WG/GduFhIQ4XOszrusBAQHGuXPnHLbPOIZVq1YZly5dMpo1a2aUKlXKOHDggG2bbdu2GZKM9957z+G169aty1Reu3Zth3N0I860hYzPVmRkpMM1Z9SoUYanp6dx4cIFW1lW18rBgwcbxYoVMxITE21l2V0Xly9fbnh4eDi0O8MwjIULFxqSjO3btzt1XIA9utsBdv773//K09NTw4cPdyh/5plnZBiGvvjii5vWYd+/PyUlRX/99ZeqVq2qoKAg7d+/P8ex/eMf/3BYfuCBB/TXX38pPj4+x3U++eSTtnlPT081aNBAhmFo4MCBtvKgoCDVqFFDv/32W6bX9+nTR/7+/rblhx9+WGXLltV///tfSdLBgwd17NgxPfbYY/rrr7/0559/6s8//1RCQoJatWqlr7/+OlN3nuuPMzv//e9/1ahRI4cuecWLF9egQYN04sSJbLuj3Mh//vMfmUwmTZgwIdO667uERUZGOtyNqFevngICAhzO0622BYvFov79+zuUZTzD8tlnnyklJcWp45KkjRs3Kjk5WcOGDXM4lpEjR970tYGBgZKk9evXO3TxdJWz760kRUREKDw83LZcsWJFde7cWevXr8/UBSc3ZZynkSNHysPj2p/Ip556SgEBAZm6jRUvXtzhmUaz2axGjRpl+XlxhqvXoGbNmqlWrVpO1x8WFqaoqCiHslWrVumBBx5QiRIlbJ/RP//8U5GRkUpLS7N1cwwKClJCQoI2bNhww/hd+Vz27dvX6Weictr2b6R///4Ozys98MADkuT0+5fTa6inp6dtv1arVX///bdSU1PVoEGDLK8N3bt3V+nSpbOM4eLFi2rTpo0OHz6sr776yuEu1KpVqxQYGKjWrVs7vLfh4eEqXrx4pu59znKmLWQYNGiQwzXngQceUFpamk6ePGkrs28Dly5d0p9//qkHHnhAV65c0eHDhx3qy+q6uGrVKtWsWVN33XWXw3G2bNlSknJ8nCjaSJIAOydPnlS5cuUc/vGXro12Z39Rz87Vq1f14osv2vr2lypVSqVLl9aFCxdsz2/kRMWKFR2WS5QoIUkO/cpvtc7AwED5+PioVKlSmcqz2k+1atUclk0mk6pWrWp7vuPYsWOS0v8RKl26tMO0aNEiJSUlZTonYWFhTsV+8uRJ1ahRI1O5K+/V9X799VeVK1dOJUuWvOm21587Kf09sT9Pt9oW7rjjjkwPnDdr1kzdu3fXpEmTVKpUKXXu3FkxMTE37XefcT6uf89Kly5ta0vZCQsLU3R0tBYtWqRSpUopKipK8+bNc7k9O/veZhWnJFWvXl1XrlzJ9DxDbso4T9e3LbPZrDvvvDNTuypfvnymBPr6duDq/l25BrlyTrPb/tixY1q3bl2mz2hkZKSk9AEmJOnpp59W9erV1a5dO5UvX14DBgzI8jk8Vz6XrsSf07Z/I7d6Xb2Va+jSpUtVr149+fj4KDg4WKVLl9bnn3+e5efqRudp5MiR2rNnjzZu3KjatWs7rDt27JguXryoMmXKZHp/L1++bHtvXeVMW8jgzDk+dOiQunbtqsDAQAUEBKh06dK2Lx+uPx9ZXRePHTumQ4cOZTrG6tWrS1KOjxNFG88kAbls2LBhiomJ0ciRIxUREaHAwECZTCY9+uijtzQiUXYjcBn//2xHdg+/3+hb96zqvNl+XJFxvDNmzMjUxz5D8eLFHZYLykhbN+PMebrVtpDVucj48cidO3fq008/1fr16zVgwADNnDlTO3fuzHQ+c8vMmTPVr18/ffzxx/ryyy81fPhwTZ06VTt37lT58uWdqiO339uctPnclpufl5xw9Zxmtb3ValXr1q01ZsyYLF+T8Y9mmTJldPDgQa1fv15ffPGFvvjiC8XExKhPnz5aunSp68FnE0928qLt3+r7l9Nr6Lvvvqt+/fqpS5cu+te//qUyZcrI09NTU6dO1a+//prptTc6T507d9bKlSs1bdo0LVu2zOEOqNVqVZkyZfTee+9l+drs7k7djCtt4Wbn48KFC2rWrJkCAgL00ksvqUqVKvLx8dH+/fs1duzYTNfK7Npw3bp1NWvWrCz3VaFChZwcJoo4kiTATqVKlbRx40ZdunTJ4ZvcjNv9lSpVspVl9w/a6tWr1bdvX4fRjBITEzM9ZJvbMr6du34/Obmj4qyMO0UZDMPQL7/8onr16km69nB8QECA7Vvp3FKpUiUdOXIkU3lW75WzqlSpovXr1+vvv/926m7SzeRlW2jcuLEaN26sV155RStWrFDv3r21cuVKh+4/9jLOx7Fjx3TnnXfayv/44w+nvzWvW7eu6tatq+eff17ffvutmjRpooULF2ry5MmSsv9M5MT1bUuSjh49qmLFitn+sStRokSW5zKrNu9sbBnn6ciRIw7nKTk5WcePH8/1dpzV/p29BuWWKlWq6PLly04dm9lsVqdOndSpUydZrVY9/fTTevPNN/XCCy+oatWqufK5vNl75WrbL4hWr16tO++8U2vWrHE43qy6+t5Mly5d1KZNG/Xr10/+/v5asGCBbV2VKlW0ceNGNWnS5KYJqauf35u1BWd99dVX+uuvv7RmzRqHwYGOHz/udB1VqlTRd999p1atWuXqdQhFG93tADvt27dXWlqa3njjDYfy1157TSaTyWG0Kj8/vyz/QfP09Mz0LeTrr7+e599uBwQEqFSpUpmGSZ4/f36e7XPZsmW6dOmSbXn16tWKjY21nafw8HBVqVJF//73v3X58uVMr7+VblPt27fX7t27tWPHDltZQkKC3nrrLVWuXNml5zQydO/eXYZhaNKkSZnW5eTOQF60hfPnz2eqM+Mu3Y26HUVGRsrb21uvv/66w+szRi67kfj4eKWmpjqU1a1bVx4eHg77zO4zkRM7duxweDbj9OnT+vjjj9WmTRvbN9NVqlTRxYsX9f3339u2i42N1dq1azPV52xskZGRMpvNmjt3rsN5Wrx4sS5evKgOHTrcwlHdnCvXoNzSs2dP7dixQ+vXr8+07sKFC7b3/q+//nJY5+HhYftCJKMd5MbnMuO30a5/v3La9guijDZsfzy7du1yOG+u6NOnj+bOnauFCxdq7NixtvKePXsqLS1NL7/8cqbXpKamOpxjVz6/zrQFZ2V1LpKTk13629WzZ0/9/vvvevvttzOtu3r1qhISElyKCZC4kwQ46NSpk1q0aKHnnntOJ06cUP369fXll1/q448/1siRIx0e1A8PD9fGjRs1a9YslStXTmFhYbrvvvvUsWNHLV++XIGBgapVq5Z27NihjRs32oaFzktPPvmkpk2bpieffFINGjTQ119/raNHj+bZ/kqWLKmmTZuqf//+iouL0+zZs1W1alU99dRTktL/cC5atEjt2rVT7dq11b9/f91xxx36/ffftWXLFgUEBOjTTz/N0b7HjRun999/X+3atdPw4cNVsmRJLV26VMePH9d//vMfhy4nzmrRooWeeOIJzZ07V8eOHVPbtm1ltVq1bds2tWjRQkOHDnWpvrxoC0uXLtX8+fPVtWtXValSRZcuXdLbb7+tgIAAtW/fPtvXlS5dWqNHj9bUqVPVsWNHtW/fXgcOHNAXX3yR6fmJ623evFlDhw5Vjx49VL16daWmpmr58uXy9PRU9+7dbdtl95nIiTp16igqKsphCHBJDgnso48+qrFjx6pr164aPny4rly5ogULFqh69eqZHn53NrbSpUtr/PjxmjRpktq2bauHHnpIR44c0fz589WwYcM8/+FpV65BueVf//qXPvnkE3Xs2FH9+vVTeHi4EhIS9MMPP2j16tU6ceKESpUqpSeffFJ///23WrZsqfLly+vkyZN6/fXXdffdd9ueOcqNz+Xdd98tT09PTZ8+XRcvXpTFYlHLli21YsWKHLX9gqhjx45as2aNunbtqg4dOuj48eNauHChatWqleUXSs4YOnSo4uPj9dxzzykwMFDPPvusmjVrpsGDB2vq1Kk6ePCg2rRpI29vbx07dkyrVq3SnDlz9PDDD0tK/4wsWLBAkydPVtWqVVWmTBnbwAfXc6YtOOv+++9XiRIl1LdvXw0fPlwmk0nLly936YupJ554Qh9++KH+8Y9/aMuWLWrSpInS0tJ0+PBhffjhh7bfBgNcko8j6QEFzvVDgBtG+nC1o0aNMsqVK2d4e3sb1apVM2bMmOEwhKlhGMbhw4eNBx980PD19XUYQvn8+fNG//79jVKlShnFixc3oqKijMOHDxuVKlW64TDL2ckYavaPP/5wKL9+6GLDSB9GdeDAgUZgYKDh7+9v9OzZ0zh37ly2Q4BfX2ffvn0NPz+/TDFcP1RxRuzvv/++MX78eKNMmTKGr6+v0aFDB+PkyZOZXn/gwAGjW7duRnBwsGGxWIxKlSoZPXv2NDZt2nTTmG7k119/NR5++GEjKCjI8PHxMRo1amR89tlnmbaTk0OAG0b60LwzZsww7rrrLsNsNhulS5c22rVrZ+zbt++m9V3/Ht9KW8hqeGjDMIz9+/cbvXr1MipWrGhYLBajTJkyRseOHR2Gy85OWlqaMWnSJKNs2bKGr6+v0bx5c+PHH3+8aTy//fabMWDAAKNKlSqGj4+PUbJkSaNFixbGxo0bHerP7jNxo/c2uyHAhwwZYrz77rtGtWrVDIvFYtxzzz1Zfla+/PJLo06dOobZbDZq1KhhvPvuu1nWmV1sWX2ODCN9yO+77rrL8Pb2NkJCQox//vOfxvnz5x22ye49ym5o8utl93pnr0GutGvDSG+fHTp0yHLdpUuXjPHjxxtVq1Y1zGazUapUKeP+++83/v3vfxvJycmGYRjG6tWrjTZt2hhlypQxzGazUbFiRWPw4MFGbGysQ13OfC7th6/Oyttvv23ceeedhqenp60t3krbz24I8Ov3f6OfUrB3q9dQq9VqTJkyxahUqZKtfX/22WeZ2k5GPDNmzMhUZ3bHMGbMGEOS8cYbb9jK3nrrLSM8PNzw9fU1/P39jbp16xpjxowxzpw5Y9vm7NmzRocOHQx/f39D0g2HA3emLWR8tq4fsj2r69327duNxo0bG76+vka5cuWMMWPGGOvXr3f6umgYhpGcnGxMnz7dqF27tmGxWIwSJUoY4eHhxqRJk4yLFy9meyxAdkyGkU9PlwIAAADAbYBnkgAAAADADkkSAAAAANghSQIAAAAAOyRJAAAAAGCHJAkAAAAA7JAkAQAAAICdQv9jslarVWfOnJG/v79MJpO7wwEAAADgJoZh6NKlSypXrtwNf+C60CdJZ86cUYUKFdwdBgAAAIAC4vTp0ypfvny26wt9kuTv7y8p/UQEBAS4ORoAty2rVYqLS58PCZFu8O0TAAAomOLj41WhQgVbjpCdQp8kZXSxCwgIIEkCkHNXr0q9e6fPb9sm+fq6Nx4AAJBjN3sMh69CAQAAAMAOSRIAAAAA2CFJAgAAAAA7hf6ZJAAAANzeDMNQamqq0tLS3B0KCjhPT095eXnd8k//kCQBAACgwEpOTlZsbKyuXLni7lBwmyhWrJjKli0rs9mc4zpIkgAAAFAgWa1WHT9+XJ6enipXrpzMZvMt3yFA4WUYhpKTk/XHH3/o+PHjqlat2g1/MPZGSJIAwBmenlKPHtfmAQB5Ljk5WVarVRUqVFCxYsXcHQ5uA76+vvL29tbJkyeVnJwsHx+fHNVDkgQAzjCbpbFj3R0FABRJOb0bgKIpN9oLLQ4AAAAA7HAnCQCcYRjShQvp80FBEn3iAQAotLiTBADOSEyUWrdOnxIT3R0NAKAQOHHihEwmkw4ePOjuUAqUypUra/bs2W6NgSQJAAAAyGX9+vWTyWSyTcHBwWrbtq2+//572zYVKlRQbGys6tSpc0v7qly5skwmk1auXJlpXe3atWUymbRkyZJb2kdOmEwmffTRRy6/bs+ePRo0aFDuB+QCkiQAAAAgD7Rt21axsbGKjY3Vpk2b5OXlpY4dO9rWe3p6KjQ0VF5et/4ETIUKFRQTE+NQtnPnTp09e1Z+fn63XH9+Kl26tNtHMyRJAgAAwO3n6tXsp+Rk57dNSnJu2xywWCwKDQ1VaGio7r77bo0bN06nT5/WH3/8ISnr7naffPKJqlWrJh8fH7Vo0UJLly6VyWTShYznYrPRu3dvbd26VadPn7aVvfPOO+rdu3emJOzUqVPq3LmzihcvroCAAPXs2VNxcXG29f369VOXLl0cXjNy5Eg1b97ctty8eXMNHz5cY8aMUcmSJRUaGqqJEyfa1leuXFmS1LVrV5lMJtvyr7/+qs6dOyskJETFixdXw4YNtXHjRod9Xd/dzmQyadGiReratauKFSumatWq6ZNPPrnh+bhVJEkAAAC4/TzwQPbTv/7luG3r1tlvO2yY47adOmW93S26fPmy3n33XVWtWlXBwcFZbnP8+HE9/PDD6tKli7777jsNHjxYzz33nFP1h4SEKCoqSkuXLpUkXblyRR988IEGDBjgsJ3ValXnzp31999/a+vWrdqwYYN+++03PfLIIy4f09KlS+Xn56ddu3bp1Vdf1UsvvaQNGzZISu8yJ0kxMTGKjY21LV++fFnt27fXpk2bdODAAbVt21adOnXSqVOnbrivSZMmqWfPnvr+++/Vvn179e7dW3///bfLMTuL0e2AXFB53Od5VveJaR3yrG4AAJB3PvvsMxUvXlySlJCQoLJly+qzzz7L9nd83nzzTdWoUUMzZsyQJNWoUUM//vijXnnlFaf2N2DAAD3zzDN67rnntHr1alWpUkV33323wzabNm3SDz/8oOPHj6tChQqSpGXLlql27dras2ePGjZs6PTx1atXTxMmTJAkVatWTW+88YY2bdqk1q1bq3Tp0pKkoKAghYaG2l5Tv3591a9f37b88ssva+3atfrkk080dOjQbPfVr18/9erVS5I0ZcoUzZ07V7t371bbtm2djtcVJEkAAAC4/Wzblv06T0/H5f+/u5Gl6xOWTz/NeUzXadGihRYsWCBJOn/+vObPn6927dpp9+7dqlSpUqbtjxw5kilJadSokdP769ChgwYPHqyvv/5a77zzTqa7SJL0888/q0KFCrYESZJq1aqloKAg/fzzzy4nSfbKli2rc+fO3fA1ly9f1sSJE/X5558rNjZWqampunr16k3vJNnvy8/PTwEBATfd161we3e733//XY8//riCg4Pl6+urunXrau/evbb1hmHoxRdfVNmyZeXr66vIyEgdO3bMjREDKJI8PaWOHdOn6//4AgDyn69v9pPZ7Py2Fotz2+aAn5+fqlatqqpVq6phw4ZatGiREhIS9Pbbb+fwoG/My8tLTzzxhCZMmKBdu3apd+/eOarHw8NDhmE4lKWkpGTaztvb22HZZDLJarXesO7Ro0dr7dq1mjJlirZt26aDBw+qbt26Sr7+ObJc2NetcGuSdP78eTVp0kTe3t764osv9NNPP2nmzJkqUaKEbZtXX31Vc+fO1cKFC7Vr1y75+fkpKipKifxOCYD8ZDZLEyemT9f/8QUAwAkmk0keHh66ms1AEDVq1HC4WSBde7bHWQMGDNDWrVvVuXNnh/+pM9SsWVOnT592GODhp59+0oULF1SrVi1J6aPLxcbGOrwuJ7/l5O3trbS0NIey7du3q1+/furatavq1q2r0NBQnThxwuW685pbk6Tp06fbhits1KiRwsLC1KZNG1WpUkVS+l2k2bNn6/nnn1fnzp1Vr149LVu2TGfOnMnRmOsAAABAfklKStLZs2d19uxZ/fzzzxo2bJguX76sTp06Zbn94MGDdfjwYY0dO1ZHjx7Vhx9+aPt9I5PJ5NQ+a9asqT///DPTcOAZIiMjVbduXfXu3Vv79+/X7t271adPHzVr1kwNGjSQJLVs2VJ79+7VsmXLdOzYMU2YMEE//vijy8dfuXJlbdq0SWfPntX58+clpT+7tGbNGh08eFDfffedHnvssTy9I5RTbk2SPvnkEzVo0EA9evRQmTJldM899zjcfjx+/LjOnj2ryMhIW1lgYKDuu+8+7dixI8s6k5KSFB8f7zABwC0zjGvDwF7XBQEAgKysW7dOZcuWVdmyZXXfffdpz549WrVqlcNQ2vbCwsK0evVqrVmzRvXq1dOCBQtso9tZru8WeAMZj7FkxWQy6eOPP1aJEiX04IMPKjIyUnfeeac++OAD2zZRUVF64YUXNGbMGDVs2FCXLl1Snz59nD/w/zdz5kxt2LBBFSpU0D333CNJmjVrlkqUKKH7779fnTp1UlRUlO69916X685rJuP6Dof5yMfHR5IUHR2tHj16aM+ePRoxYoQWLlyovn376ttvv1WTJk105swZlS1b1va6nj17ymQyObyZGSZOnKhJkyZlKr948aICAgLy7mCKgLwcwU26vUdxY3S7IuDq1WtDwG7bluP+6QAA5yUmJur48eMKCwuz/d9Y1LzyyitauHChQ/c43NiN2k18fLwCAwNvmhu4dXQ7q9WqBg0aaMqUKZKke+65Rz/++KMtScqJ8ePHKzo62rYcHx/vMHoHAAAAUFDNnz9fDRs2VHBwsLZv364ZM2bccGhs5A23Jklly5a1PSCWoWbNmvrPf/4jSbYx1ePi4hzuJMXFxWUa8z2DxWJx6XYkAAAAUFAcO3ZMkydP1t9//62KFSvqmWee0fjx490dVpHj1meSmjRpoiNHjjiUHT161DZufFhYmEJDQ7Vp0ybb+vj4eO3atUsRERH5GisAAACQ11577TWdOXNGiYmJOnr0qF544QV5efHTpvnNrWd81KhRuv/++zVlyhT17NlTu3fv1ltvvaW33npLUvqDZSNHjtTkyZNVrVo1hYWF6YUXXlC5cuXUpUsXd4YOAAAAoJBya5LUsGFDrV27VuPHj9dLL72ksLAwzZ492+GHr8aMGaOEhAQNGjRIFy5cUNOmTbVu3boi+/AeAABAUePGccZwG8qN9uL2e3cdO3ZUx44ds11vMpn00ksv6aWXXsrHqAAAAOBu3t7ekqQrV65kO6Q1cL0rV65IutZ+csLtSRIA3BY8PaVWra7NAwDynKenp4KCgnTu3DlJUrFixZz+UVUUPYZh6MqVKzp37pyCgoLkeQt/r0mSAMAZZrM0fbq7owCAIidjtOOMRAm4maCgIFu7ySmSJAAAABRYJpNJZcuWVZkyZZSSkuLucFDAeXt739IdpAwkSQAAACjwPD09c+WfX8AZbv2dJAC4bVy9KjVokD5dveruaAAAQB4iSQIAAAAAOyRJAAAAAGCHJAkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSBAAAAAB2+J0kAHCGp6fUpMm1eQAAUGiRJAGAM8xmac4cd0cBAADyAd3tAAAAAMAOSRIAAAAA2CFJAgBnXL0qNW2aPl296u5oAABAHuKZJABwVmKiuyMAAAD5gDtJAAAAAGCHJAkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSBAAAAAB2GN0OAJzh4SHde++1eQAAUGiRJAGAMywW6a233B0FAADIB3wdCgAAAAB2SJIAAAAAwA5JEgA44+pVKTIyfbp61d3RAACAPMQzSQDgrAsX3B0BAADIB9xJAgAAAAA7JEkAAAAAYIckCQAAAADskCQBAAAAgB2SJAAAAACww+h2AOAMDw+pVq1r8wAAoNAiSQIAZ1gs0rJl7o4CAADkA74OBQAAAAA7JEkAAAAAYIckCQCckZgodeqUPiUmujsaAACQh3gmCQCcYRhSbOy1eQAAUGhxJwkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSBAAAAAB2SJIAAAAAwA6j2wGAM0wm6c47r80DAIBCiyQJAJzh4yN9+KG7owAAAPmA7nYAAAAAYIckCQAAAADskCQBgDMSE6WePdOnxER3RwMAAPKQW5OkiRMnymQyOUx33XWXbX1iYqKGDBmi4OBgFS9eXN27d1dcXJwbIwZQZBmG9Ntv6ZNhuDsaAACQh9x+J6l27dqKjY21Td98841t3ahRo/Tpp59q1apV2rp1q86cOaNu3bq5MVoAAAAAhZ3bR7fz8vJSaGhopvKLFy9q8eLFWrFihVq2bClJiomJUc2aNbVz5041btw4v0MFAAAAUAS4/U7SsWPHVK5cOd15553q3bu3Tp06JUnat2+fUlJSFBkZadv2rrvuUsWKFbVjx45s60tKSlJ8fLzDBAAAAADOcmuSdN9992nJkiVat26dFixYoOPHj+uBBx7QpUuXdPbsWZnNZgUFBTm8JiQkRGfPns22zqlTpyowMNA2VahQIY+PAgAAAEBh4tbudu3atbPN16tXT/fdd58qVaqkDz/8UL6+vjmqc/z48YqOjrYtx8fHkygBAAAAcJrbn0myFxQUpOrVq+uXX35R69atlZycrAsXLjjcTYqLi8vyGaYMFotFFoslH6IFUKSYTFLZstfmAQBAoeX2Z5LsXb58Wb/++qvKli2r8PBweXt7a9OmTbb1R44c0alTpxQREeHGKAEUST4+0qefpk8+Pu6OBgAA5CG33kkaPXq0OnXqpEqVKunMmTOaMGGCPD091atXLwUGBmrgwIGKjo5WyZIlFRAQoGHDhikiIoKR7QAAAADkGbcmSf/73//Uq1cv/fXXXypdurSaNm2qnTt3qnTp0pKk1157TR4eHurevbuSkpIUFRWl+fPnuzNkAAAAAIWcW5OklStX3nC9j4+P5s2bp3nz5uVTRACQjaQk6amn0uffflvi2UcAAAqtAjVwAwAUWFar9NNP1+YBAEChVaAGbgAAAAAAdyNJAgAAAAA7JEkAAAAAYIckCQAAAADsMHADCozK4z7Ps7pPTOuQZ3UDAACgcCFJAgBnBQW5OwIAAJAPSJIAwBm+vtLGje6OAgAA5AOeSQIAAAAAOyRJAAAAAGCHJAkAnJGUJA0alD4lJbk7GgAAkId4JgkAnGG1Svv3X5sHAACFFneSAAAAAMAOSRIAAAAA2CFJAgAAAAA7JEkAAAAAYIckCQAAAADsMLodADjLx8fdEQAAgHxAkgQAzvD1lb75xt1RAACAfEB3OwAAAACwQ5IEAAAAAHZIkgDAGcnJ0ogR6VNysrujAQAAeYhnkgDAGWlp0vbt1+YBAEChxZ0kAAAAALBDkgQAAAAAdkiSAAAAAMAOSRIAAAAA2CFJAgAAAAA7JEkAAAAAYIchwAHAGb6+0t697o4CAADkA+4kAQAAAIAdkiQAAAAAsEOSBADOSE6Wxo5Nn5KT3R0NAADIQyRJAOCMtDRp06b0KS3N3dEAAIA8RJIEAAAAAHZIkgAAAADADkkSAAAAANghSQIAAAAAOyRJAAAAAGCHJAkAAAAA7Hi5OwAAuC34+Ejbtl2bBwAAhRZJEgA4w2SSfH3dHQUAAMgHdLcDAAAAADvcSQIAZyQnS1OmpM8/+6xkNrs3HgAAkGe4kwQAzkhLkz77LH1KS3N3NAAAIA+RJAEAAACAHZIkAAAAALBDkgQAAAAAdkiSAAAAAMBOgUmSpk2bJpPJpJEjR9rKEhMTNWTIEAUHB6t48eLq3r274uLi3BckAAAAgEKvQCRJe/bs0Ztvvql69eo5lI8aNUqffvqpVq1apa1bt+rMmTPq1q2bm6IEAAAAUBS4PUm6fPmyevfurbffflslSpSwlV+8eFGLFy/WrFmz1LJlS4WHhysmJkbffvutdu7c6caIARRJPj7Shg3pk4+Pu6MBAAB5yO1J0pAhQ9ShQwdFRkY6lO/bt08pKSkO5XfddZcqVqyoHTt2ZFtfUlKS4uPjHSYAuGUmk1SiRPpkMrk7GgAAkIe83LnzlStXav/+/dqzZ0+mdWfPnpXZbFZQUJBDeUhIiM6ePZttnVOnTtWkSZNyO1QAAAAARYTb7iSdPn1aI0aM0HvvvSefXOy6Mn78eF28eNE2nT59OtfqBlCEJSdL06enT8nJ7o4GAADkIbclSfv27dO5c+d07733ysvLS15eXtq6davmzp0rLy8vhYSEKDk5WRcuXHB4XVxcnEJDQ7Ot12KxKCAgwGECgFuWliatWpU+paW5OxoAAJCH3NbdrlWrVvrhhx8cyvr376+77rpLY8eOVYUKFeTt7a1Nmzape/fukqQjR47o1KlTioiIcEfIAAAAAIoAtyVJ/v7+qlOnjkOZn5+fgoODbeUDBw5UdHS0SpYsqYCAAA0bNkwRERFq3LixO0IGAAAAUAS4deCGm3nttdfk4eGh7t27KykpSVFRUZo/f767wwIAAABQiBWoJOmrr75yWPbx8dG8efM0b9489wQEAAAAoMhx++8kAQAAAEBBQpIEAAAAAHYKVHc7ACiwLBbpk0+uzQMAgEKLJAkAnOHhIZUr5+4oAABAPqC7HQAAAADY4U4SADgjJUXK+AmCp5+WvL3dGw8AAMgz3EkCAGekpkrLl6dPqanujgYAAOQhkiQAAAAAsEOSBAAAAAB2XE6Sfvvtt7yIAwAAAAAKBJeTpKpVq6pFixZ69913lZiYmBcxAQAAAIDbuDy63f79+xUTE6Po6GgNHTpUjzzyiAYOHKhGjRrlRXxArqg87nN3hwAAAIDbhMt3ku6++27NmTNHZ86c0TvvvKPY2Fg1bdpUderU0axZs/THH3/kRZwAAAAAkC9yPHCDl5eXunXrplWrVmn69On65ZdfNHr0aFWoUEF9+vRRbGxsbsYJAO5lsUgffpg+WSzujgYAAOShHCdJe/fu1dNPP62yZctq1qxZGj16tH799Vdt2LBBZ86cUefOnXMzTgBwLw8P6c470ycPBgYFAKAwc/mZpFmzZikmJkZHjhxR+/bttWzZMrVv314e//9PQ1hYmJYsWaLKlSvndqwAAAAAkOdcTpIWLFigAQMGqF+/fipbtmyW25QpU0aLFy++5eAAoMBISZFiYtLn+/eXvL3dGw8AAMgzLidJx44du+k2ZrNZffv2zVFAAFAgpaZKb72VPv/EEyRJAAAUYi53rI+JidGqVasyla9atUpLly7NlaAAAAAAwF1cTpKmTp2qUqVKZSovU6aMpkyZkitBAQAAAIC7uJwknTp1SmFhYZnKK1WqpFOnTuVKUAAAAADgLi4nSWXKlNH333+fqfy7775TcHBwrgQFAAAAAO7icpLUq1cvDR8+XFu2bFFaWprS0tK0efNmjRgxQo8++mhexAgAAAAA+cbl0e1efvllnThxQq1atZKXV/rLrVar+vTpwzNJAAAAAG57LidJZrNZH3zwgV5++WV999138vX1Vd26dVWpUqW8iA8ACgaLRVq27No8AAAotFxOkjJUr15d1atXz81YAKDg8vCQatVydxQAACAfuJwkpaWlacmSJdq0aZPOnTsnq9XqsH7z5s25FhwAAAAA5DeXk6QRI0ZoyZIl6tChg+rUqSOTyZQXcQFAwZKSIr3/fvp8r16St7d74wEAAHnG5SRp5cqV+vDDD9W+ffu8iAcACqbUVGnu3PT5Hj1IkgAAKMRcHgLcbDaratWqeRELAAAAALidy0nSM888ozlz5sgwjLyIBwAAAADcyuXudt988422bNmiL774QrVr15b3dV1O1qxZk2vBAQAAAEB+czlJCgoKUteuXfMiFgAAAABwO5eTpJiYmLyIAwAAAAAKBJefSZKk1NRUbdy4UW+++aYuXbokSTpz5owuX76cq8EBAAAAQH5z+U7SyZMn1bZtW506dUpJSUlq3bq1/P39NX36dCUlJWnhwoV5EScAuJfFIr355rV5AABQaLl8J2nEiBFq0KCBzp8/L19fX1t5165dtWnTplwNDgAKDA8PKTw8ffLI0U14AABwm3D5TtK2bdv07bffymw2O5RXrlxZv//+e64FBgAAAADu4HKSZLValZaWlqn8f//7n/z9/XMlKAAocFJTpYyfOOjWTfJy+fIJAABuEy73GWnTpo1mz55tWzaZTLp8+bImTJig9u3b52ZsAFBwpKRIr76aPqWkuDsaAACQh1z+KnTmzJmKiopSrVq1lJiYqMcee0zHjh1TqVKl9P777+dFjAAAAACQb1xOksqXL6/vvvtOK1eu1Pfff6/Lly9r4MCB6t27t8NADgAAAABwO8pRp3ovLy89/vjjuR0LAAAAALidy0nSsmXLbri+T58+OQ4GAAAAANzN5SRpxIgRDsspKSm6cuWKzGazihUrRpIEAAAA4Lbm8uh258+fd5guX76sI0eOqGnTpgzcAAAAAOC2lys/9FGtWjVNmzZNjz/+uA4fPpwbVQJAwWI2Sxk/f3Ddj2kDAIDCJdd+DdHLy0tnzpzJreoAFAKVx32eZ3WfmNYhz+rOkqen1LRp/u4TAAC4hctJ0ieffOKwbBiGYmNj9cYbb6hJkya5FhgAAAAAuIPLSVKXLl0clk0mk0qXLq2WLVtq5syZuRUXABQsqanSF1+kz7drJ3nl2o14AABQwLg8cIPVanWY0tLSdPbsWa1YsUJly5Z1qa4FCxaoXr16CggIUEBAgCIiIvRFxj8hkhITEzVkyBAFBwerePHi6t69u+Li4lwNGQBuXUqKNGlS+pSS4u5oAABAHnI5ScpN5cuX17Rp07Rv3z7t3btXLVu2VOfOnXXo0CFJ0qhRo/Tpp59q1apV2rp1q86cOaNu3bq5M2QAAAAAhZzL/UWio6Od3nbWrFk3XN+pUyeH5VdeeUULFizQzp07Vb58eS1evFgrVqxQy5YtJUkxMTGqWbOmdu7cqcaNG2dZZ1JSkpKSkmzL8fHxTscLAAAAAC4nSQcOHNCBAweUkpKiGjVqSJKOHj0qT09P3XvvvbbtTCaTS/WmpaVp1apVSkhIUEREhPbt26eUlBRFRkbatrnrrrtUsWJF7dixI9skaerUqZo0aZKrhwUAAAAAknKQJHXq1En+/v5aunSpSpQoISn9B2b79++vBx54QM8884xL9f3www+KiIhQYmKiihcvrrVr16pWrVo6ePCgzGazgoKCHLYPCQnR2bNns61v/PjxDne74uPjVaFCBZdiAgAAAFB0uZwkzZw5U19++aUtQZKkEiVKaPLkyWrTpo3LSVKNGjV08OBBXbx4UatXr1bfvn21detWV8OysVgsslgsOX49AAAAgKLN5SQpPj5ef/zxR6byP/74Q5cuXXI5ALPZrKpVq0qSwsPDtWfPHs2ZM0ePPPKIkpOTdeHCBYe7SXFxcQoNDXV5PwAAAADgDJdHt+vatav69++vNWvW6H//+5/+97//6T//+Y8GDhyYKyPPWa1WJSUlKTw8XN7e3tq0aZNt3ZEjR3Tq1ClFRETc8n4AwCVmszRtWvpkNrs7GgAAkIdcvpO0cOFCjR49Wo899phS/v+3Qry8vDRw4EDNmDHDpbrGjx+vdu3aqWLFirp06ZJWrFihr776SuvXr1dgYKAGDhyo6OholSxZUgEBARo2bJgiIiKyHbQBAPKMp6dkN5AMAAAovFxOkooVK6b58+drxowZ+vXXXyVJVapUkZ+fn8s7P3funPr06aPY2FgFBgaqXr16Wr9+vVq3bi1Jeu211+Th4aHu3bsrKSlJUVFRmj9/vsv7AQAAAABnuZwkZYiNjVVsbKwefPBB+fr6yjAMl4f9Xrx48Q3X+/j4aN68eZo3b15OwwSA3JGWJm3Zkj7fokX6nSUAAFAouZwk/fXXX+rZs6e2bNkik8mkY8eO6c4779TAgQNVokQJzZw5My/iBAD3Sk6Wxo1Ln9+2TfL1dW88AAAgz7g8cMOoUaPk7e2tU6dOqVixYrbyRx55ROvWrcvV4AAAAAAgv7l8J+nLL7/U+vXrVb58eYfyatWq6eTJk7kWGAAAAAC4g8t3khISEhzuIGX4+++/+RFXAAAAALc9l5OkBx54QMuWLbMtm0wmWa1Wvfrqq2rRokWuBgcAAAAA+c3l7navvvqqWrVqpb179yo5OVljxozRoUOH9Pfff2v79u15ESMAAAAA5BuX7yTVqVNHR48eVdOmTdW5c2clJCSoW7duOnDggKpUqZIXMQIAAABAvnHpTlJKSoratm2rhQsX6rnnnsurmACg4PH2liZMuDYPAAAKLZeSJG9vb33//fd5FQsAFFxeXlKnTu6OAgAA5AOXu9s9/vjjWrx4cV7EAgAAAABu5/LADampqXrnnXe0ceNGhYeHy8/Pz2H9rFmzci04ACgw0tKkHTvS5yMiJE9P98YDAADyjFNJ0vfff686derIw8NDP/74o+69915J0tGjRx22M5lMuR8hABQEycnSyJHp89u2Sb6+bg0HAADkHaeSpHvuuUexsbEqU6aMTp48qT179ig4ODivYwMAAACAfOfUM0lBQUE6fvy4JOnEiROyWq15GhQAAAAAuItTd5K6d++uZs2aqWzZsjKZTGrQoIE8s+mP/9tvv+VqgAAAAACQn5xKkt566y1169ZNv/zyi4YPH66nnnpK/v7+eR0bAAAAAOQ7p0e3a9u2rSRp3759GjFiBEkSAAAAgELJ5SHAY2Ji8iIOAAAAACgQXE6SAKBI8vaWxoy5Ng8AAAotkiQAcIaXl9Szp7ujAAAA+cCpIcABAAAAoKjgThIAOMNqlQ4cSJ+/5x7Jg++YAAAorEiSAMAZSUnS4MHp89u2Sb6+7o0HAADkGb4KBQAAAAA7JEkAAAAAYIckCQAAAADskCQBAAAAgB2SJAAAAACwQ5IEAAAAAHYYAhwAnOHlJQ0ffm0eAAAUWvylBwBneHtLffq4OwoAAJAP6G4HAAAAAHa4kwQAzrBapcOH0+fvukvy4DsmAAAKK5IkAHBGUtK17nbbtkm+vu6NBwAA5Bm+CgUAAAAAOyRJAAAAAGCHJAkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSBAAAAAB2GAIcAJzh5SUNGnRtHgAAFFr8pQcAZ3h7X0uSAABAoUZ3OwAAAACww50kAHCG1SqdOJE+X7my5MF3TAAAFFYkSQDgjKQkqWfP9Plt2yRfX/fGAwAA8gxfhQIAAACAHe4kAUVc5XGfuzsEAACAAoU7SQAAAABghyQJAAAAAOy4NUmaOnWqGjZsKH9/f5UpU0ZdunTRkSNHHLZJTEzUkCFDFBwcrOLFi6t79+6Ki4tzU8QAAAAACju3Jklbt27VkCFDtHPnTm3YsEEpKSlq06aNEhISbNuMGjVKn376qVatWqWtW7fqzJkz6tatmxujBgAAAFCYuXXghnXr1jksL1myRGXKlNG+ffv04IMP6uLFi1q8eLFWrFihli1bSpJiYmJUs2ZN7dy5U40bN3ZH2ACKIi8v6Yknrs0DAIBCq0D9pb948aIkqWTJkpKkffv2KSUlRZGRkbZt7rrrLlWsWFE7duzIMklKSkpSUlKSbTk+Pj6PowZQJHh7SyNGuDsKAACQDwrMwA1Wq1UjR45UkyZNVKdOHUnS2bNnZTabFRQU5LBtSEiIzp49m2U9U6dOVWBgoG2qUKFCXocOAAAAoBApMEnSkCFD9OOPP2rlypW3VM/48eN18eJF23T69OlcihBAkWa1SmfOpE9Wq7ujAQAAeahAdLcbOnSoPvvsM3399dcqX768rTw0NFTJycm6cOGCw92kuLg4hYaGZlmXxWKRxWLJ65ABFDVJSdJDD6XPb9sm+fq6Nx4AAJBn3HonyTAMDR06VGvXrtXmzZsVFhbmsD48PFze3t7atGmTrezIkSM6deqUIiIi8jtcAAAAAEWAW+8kDRkyRCtWrNDHH38sf39/23NGgYGB8vX1VWBgoAYOHKjo6GiVLFlSAQEBGjZsmCIiIhjZDgAAAECecGuStGDBAklS8+bNHcpjYmLUr18/SdJrr70mDw8Pde/eXUlJSYqKitL8+fPzOVIAAAAARYVbkyTDMG66jY+Pj+bNm6d58+blQ0QAAAAAiroCM7odAAAAABQEJEkAAAAAYKdADAEOAAWep6fUo8e1eQAAUGiRJAGAM8xmaexYd0cBAADyAd3tAAAAAMAOd5IAwBmGIV24kD4fFCSZTO6MBgAA5CGSJABwRmKi1Lp1+vy2bZKvr3vjAQAAeYbudgAAAABghyQJAAAAAOyQJAEAAACAHZIkAAAAALBDkgQAAAAAdkiSAAAAAMAOQ4ADgDM8PaWOHa/NAwCAQoskCQCcYTZLEye6OwoAAJAP6G4HAAAAAHa4kwQAzjAMKTExfd7HRzKZ3BsPAADIMyRJQAFXedzn7g4BUnqC9MAD6fPbtkm+vu6NBwAA5Bm62wEAAACAHZIkAAAAALBDkgQAAAAAdkiSAAAAAMAOSRIAAAAA2CFJAgAAAAA7DAEOAM7w9JRatbo2DwAACi2SJABwhtksTZ/u7igAAEA+oLsdAAAAANghSQIAAAAAOyRJAOCMq1elBg3Sp6tX3R0NAADIQyRJAAAAAGCHJAkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSBAAAAAB2SJIAAAAAwI6XuwMAgNuCp6fUpMm1eQAAUGiRJAGAM8xmac4cd0cBAADyAd3tAAAAAMAOSRIAAAAA2CFJAgBnXL0qNW2aPl296u5oAABAHuKZJABwVmKiuyMAAAD5gCQJALJQedznDsuWlCSt+v2iJKnH818oyduS47pPTOtwS7EBAIC8RXc7AAAAALBDkgQAAAAAdkiSAAAAAMAOSRIAAAAA2GHgBgBwgmHy0I8hVWzzAACg8CJJAgAnJHt569m2w9wdBgAAyAd8HQoAAAAAdtyaJH399dfq1KmTypUrJ5PJpI8++shhvWEYevHFF1W2bFn5+voqMjJSx44dc0+wAAAAAIoEtyZJCQkJql+/vubNm5fl+ldffVVz587VwoULtWvXLvn5+SkqKkqJ/Oo9gHxmSUnSuyuf07srn5MlJcnd4QAAgDzk1meS2rVrp3bt2mW5zjAMzZ49W88//7w6d+4sSVq2bJlCQkL00Ucf6dFHH83PUAFAAUkJ7g4BAADkgwL7TNLx48d19uxZRUZG2soCAwN13333aceOHdm+LikpSfHx8Q4TAAAAADirwCZJZ8+elSSFhIQ4lIeEhNjWZWXq1KkKDAy0TRUqVMjTOAEAAAAULgU2Scqp8ePH6+LFi7bp9OnT7g4JAAAAwG2kwCZJoaGhkqS4uDiH8ri4ONu6rFgsFgUEBDhMAAAAAOCsApskhYWFKTQ0VJs2bbKVxcfHa9euXYqIiHBjZAAAAAAKM7eObnf58mX98ssvtuXjx4/r4MGDKlmypCpWrKiRI0dq8uTJqlatmsLCwvTCCy+oXLly6tKli/uCBlAkGSYPHQuuYJsHAACFl1uTpL1796pFixa25ejoaElS3759tWTJEo0ZM0YJCQkaNGiQLly4oKZNm2rdunXy8fFxV8gAiqhkL2890/EZd4cBAADygVuTpObNm8swjGzXm0wmvfTSS3rppZfyMarbW+Vxn7s7BAAAAOC2Rp8RAAAAALDj1jtJAHC7sKQma95HUyVJQ7qMV5KX2c0RAQCAvEKSBADOMAyVSThvmwcAAIUX3e0AAAAAwA5JEgAAAADYIUkCAAAAADskSQAAAABghyQJAAAAAOwwuh0AOMNk0qnAUNs8AAAovEiSAMAJSV5mDe0yzt1hAACAfEB3OwAAAACwQ5IEAAAAAHbobgcATrCkJmvmZ7MkSc90jFaSl9nNEQEAgLxCkgQAzjAMVbx41jYPAAAKL7rbAQAAAIAdkiQAAAAAsEOSBAAAAAB2eCYJwG2p8rjP3R0CAAAopLiTBAAAAAB2uJMEAM4wmXTOr4RtHgAAFF4kSQDghCQvs558eIK7wwAAAPmA7nYAAAAAYIckCQAAAADs0N0OAJxgTk3R1HVzJUnj2w5Xspd3juvK65H5TkzrkKf1AwBQ2JEkAYATTIZV1f46bZsHAACFF93tAAAAAMAOSRIAAAAA2CFJAgAAAAA7JEkAAAAAYIeBGwAABUZejvzHqH8AAGeRJAGAk+Itfu4OAQAA5AOSJABwQpK3RY8/+oq7wwAAAPmAZ5IAAAAAwA5JEgAAAADYobsdADjBnJqiiRsXSpImRv5DyV7ebo4IAADkFZKkfJaXIzcByDsmw6o6cb/a5gEAQOFFdzsAAAAAsEOSBAAAAAB2SJIAAAAAwA5JEgAAAADYIUkCAAAAADuMbgcATkryNLs7BLdjhM7scW6ydmJaB3eHAMAJeX0Nu92uBSRJAOCEJG+Lejz+qrvDAAAA+YDudgAAAABghyQJAAAAAOzQ3Q4AnOCdlqJnt7wjSZrSYoBSPL3dHBEAAMgrJEkA4AQPq1Xhv/9sm5enmwMCAAB5hiQJAAoZRlnLGufFPW7n857Xo3Hl5bkh9uzdzrEj//BMEgAAAADYuS2SpHnz5qly5cry8fHRfffdp927d7s7JAAAAACFVIFPkj744ANFR0drwoQJ2r9/v+rXr6+oqCidO3fO3aEBAAAAKIQKfJI0a9YsPfXUU+rfv79q1aqlhQsXqlixYnrnnXfcHRoAAACAQqhAD9yQnJysffv2afz48bYyDw8PRUZGaseOHVm+JikpSUlJSbblixcvSpLi4+PzNlgnWZOuuDsEADmQlpKky1Zr+nzSFVmtaW6OCEBey+v/HfLyfwJiz97tHHteyuv/UQvKucmIwzCMG25XoJOkP//8U2lpaQoJCXEoDwkJ0eHDh7N8zdSpUzVp0qRM5RUqVMiTGAEUHU0yZuY94c4wAOSTwNnujiDniN09bufY81pBOzeXLl1SYGBgtusLdJKUE+PHj1d0dLRt2Wq16u+//1ZwcLBMJpMbI0vPXCtUqKDTp08rICDArbEAN0N7xe2E9orbCe0Vt4vC2FYNw9ClS5dUrly5G25XoJOkUqVKydPTU3FxcQ7lcXFxCg0NzfI1FotFFovFoSwoKCivQsyRgICAQtPQUPjRXnE7ob3idkJ7xe2isLXVG91BylCgB24wm80KDw/Xpk2bbGVWq1WbNm1SRESEGyMDAAAAUFgV6DtJkhQdHa2+ffuqQYMGatSokWbPnq2EhAT179/f3aEBAAAAKIQKfJL0yCOP6I8//tCLL76os2fP6u6779a6desyDeZwO7BYLJowYUKm7oBAQUR7xe2E9orbCe0Vt4ui3FZNxs3GvwMAAACAIqRAP5MEAAAAAPmNJAkAAAAA7JAkAQAAAIAdkiQAAAAAsEOSlI/mzZunypUry8fHR/fdd592797t7pAAff311+rUqZPKlSsnk8mkjz76yGG9YRh68cUXVbZsWfn6+ioyMlLHjh1zT7Ao0qZOnaqGDRvK399fZcqUUZcuXXTkyBGHbRITEzVkyBAFBwerePHi6t69e6YfJAfyw4IFC1SvXj3bj3BGREToiy++sK2nraKgmjZtmkwmk0aOHGkrK4rtlSQpn3zwwQeKjo7WhAkTtH//ftWvX19RUVE6d+6cu0NDEZeQkKD69etr3rx5Wa5/9dVXNXfuXC1cuFC7du2Sn5+foqKilJiYmM+RoqjbunWrhgwZop07d2rDhg1KSUlRmzZtlJCQYNtm1KhR+vTTT7Vq1Spt3bpVZ86cUbdu3dwYNYqq8uXLa9q0adq3b5/27t2rli1bqnPnzjp06JAk2ioKpj179ujNN99UvXr1HMqLZHs1kC8aNWpkDBkyxLaclpZmlCtXzpg6daobowIcSTLWrl1rW7ZarUZoaKgxY8YMW9mFCxcMi8VivP/++26IELjm3LlzhiRj69athmGkt01vb29j1apVtm1+/vlnQ5KxY8cOd4UJ2JQoUcJYtGgRbRUF0qVLl4xq1aoZGzZsMJo1a2aMGDHCMIyie23lTlI+SE5O1r59+xQZGWkr8/DwUGRkpHbs2OHGyIAbO378uM6ePevQdgMDA3XffffRduF2Fy9elCSVLFlSkrRv3z6lpKQ4tNe77rpLFStWpL3CrdLS0rRy5UolJCQoIiKCtooCaciQIerQoYNDu5SK7rXVy90BFAV//vmn0tLSFBIS4lAeEhKiw4cPuykq4ObOnj0rSVm23Yx1gDtYrVaNHDlSTZo0UZ06dSSlt1ez2aygoCCHbWmvcJcffvhBERERSkxMVPHixbV27VrVqlVLBw8epK2iQFm5cqX279+vPXv2ZFpXVK+tJEkAgNvOkCFD9OOPP+qbb75xdyhAtmrUqKGDBw/q4sWLWr16tfr27autW7e6OyzAwenTpzVixAht2LBBPj4+7g6nwKC7XT4oVaqUPD09M40CEhcXp9DQUDdFBdxcRvuk7aIgGTp0qD777DNt2bJF5cuXt5WHhoYqOTlZFy5ccNie9gp3MZvNqlq1qsLDwzV16lTVr19fc+bMoa2iQNm3b5/OnTune++9V15eXvLy8tLWrVs1d+5ceXl5KSQkpEi2V5KkfGA2mxUeHq5NmzbZyqxWqzZt2qSIiAg3RgbcWFhYmEJDQx3abnx8vHbt2kXbRb4zDENDhw7V2rVrtXnzZoWFhTmsDw8Pl7e3t0N7PXLkiE6dOkV7RYFgtVqVlJREW0WB0qpVK/3www86ePCgbWrQoIF69+5tmy+K7ZXudvkkOjpaffv2VYMGDdSoUSPNnj1bCQkJ6t+/v7tDQxF3+fJl/fLLL7bl48eP6+DBgypZsqQqVqyokSNHavLkyapWrZrCwsL0wgsvqFy5curSpYv7gkaRNGTIEK1YsUIff/yx/P39bX3hAwMD5evrq8DAQA0cOFDR0dEqWbKkAgICNGzYMEVERKhx48Zujh5Fzfjx49WuXTtVrFhRly5d0ooVK/TVV19p/fr1tFUUKP7+/rZnOzP4+fkpODjYVl4k26u7h9crSl5//XWjYsWKhtlsNho1amTs3LnT3SEBxpYtWwxJmaa+ffsahpE+DPgLL7xghISEGBaLxWjVqpVx5MgR9waNIimrdirJiImJsW1z9epV4+mnnzZKlChhFCtWzOjatasRGxvrvqBRZA0YMMCoVKmSYTabjdKlSxutWrUyvvzyS9t62ioKMvshwA2jaLZXk2EYhpvyMwAAAAAocHgmCQAAAADskCQBAAAAgB2SJAAAAACwQ5IEAAAAAHZIkgAAAADADkkSAAAAANghSQIAAAAAOyRJAAAAAGCHJAkA4DSTyaSPPvrI3WFkqXnz5ho5cqS7w7AxDEODBg1SyZIlZTKZdPDgQZfryI1jWrJkiYKCgm6pDgAoakiSAOA2ZDKZbjhNnDgx29eeOHEix/+0w3nr1q3TkiVL9Nlnnyk2NlZ16tRxSxyPPPKIjh496pZ9A8DtysvdAQAAXBcbG2ub/+CDD/Tiiy/qyJEjtrLixYu7I6xCJy0tTSaTSR4ern+n+Ouvv6ps2bK6//778yAy5/n6+srX1zfb9cnJyTKbzfkYEQAUfNxJAoDbUGhoqG0KDAyUyWSyLZcpU0azZs1S+fLlZbFYdPfdd2vdunW214aFhUmS7rnnHplMJjVv3lyStGfPHrVu3VqlSpVSYGCgmjVrpv3797sUV/PmzTV8+HCNGTNGJUuWVGhoqMNdrazuYl24cEEmk0lfffWVJOmrr76SyWTS+vXrdc8998jX11ctW7bUuXPn9MUXX6hmzZoKCAjQY489pitXrjjsPzU1VUOHDlVgYKBKlSqlF154QYZh2NYnJSVp9OjRuuOOO+Tn56f77rvPtl/pWte0Tz75RLVq1ZLFYtGpU6eyPNatW7eqUaNGslgsKlu2rMaNG6fU1FRJUr9+/TRs2DCdOnVKJpNJlStXzvacbd++Xc2bN1exYsVUokQJRUVF6fz587b1Vqs12/MpSbNmzVLdunXl5+enChUq6Omnn9bly5czHVOGiRMn6u6779aiRYsUFhYmHx8fSdLq1atVt25d+fr6Kjg4WJGRkUpISMg2bgAozEiSAKCQmTNnjmbOnKl///vf+v777xUVFaWHHnpIx44dkyTt3r1bkrRx40bFxsZqzZo1kqRLly6pb9+++uabb7Rz505Vq1ZN7du316VLl1za/9KlS+Xn56ddu3bp1Vdf1UsvvaQNGza4fBwTJ07UG2+8oW+//VanT59Wz549NXv2bK1YsUKff/65vvzyS73++uuZ9u3l5aXdu3drzpw5mjVrlhYtWmRbP3ToUO3YsUMrV67U999/rx49eqht27a2cyNJV65c0fTp07Vo0SIdOnRIZcqUyRTb77//rvbt26thw4b67rvvtGDBAi1evFiTJ0+WlP4evPTSSypfvrxiY2O1Z8+eLI/x4MGDatWqlWrVqqUdO3bom2++UadOnZSWlub0+fTw8NDcuXN16NAhLV26VJs3b9aYMWNueG5/+eUX/ec//9GaNWt08OBBxcbGqlevXhowYIB+/vlnffXVV+rWrZtDggkARYoBALitxcTEGIGBgbblcuXKGa+88orDNg0bNjSefvppwzAM4/jx44Yk48CBAzesNy0tzfD39zc+/fRTW5kkY+3atdm+plmzZkbTpk0z7Xvs2LHZ7vv8+fOGJGPLli2GYRjGli1bDEnGxo0bbdtMnTrVkGT8+uuvtrLBgwcbUVFRDvuuWbOmYbVabWVjx441atasaRiGYZw8edLw9PQ0fv/9d4f4WrVqZYwfP94wjPRzKck4ePDgjU6N8eyzzxo1atRw2Ne8efOM4sWLG2lpaYZhGMZrr71mVKpU6Yb19OrVy2jSpEm26292PrOyatUqIzg42LZ8ffuYMGGC4e3tbZw7d85Wtm/fPkOSceLEiRvGCwBFBXeSAKAQiY+P15kzZ9SkSROH8iZNmujnn3++4Wvj4uL01FNPqVq1agoMDFRAQIAuX76cbXez7NSrV89huWzZsjp37pxLdVxfT0hIiIoVK6Y777zToez6ehs3biyTyWRbjoiI0LFjx5SWlqYffvhBaWlpql69uooXL26btm7dql9//dX2GrPZnOkYrvfzzz8rIiLCYV9NmjTR5cuX9b///c/pY8y4k3QjNzufGzduVKtWrXTHHXfI399fTzzxhP76669MXRHtVapUSaVLl7Yt169fX61atVLdunXVo0cPvf322w5d/gCgqGHgBgCAJKlv377666+/NGfOHFWqVEkWi0URERFKTk52qR5vb2+HZZPJJKvVKkm2ARAMu25cKSkpN63HZDLdsF5nXL58WZ6entq3b588PT0d1tkPdOHr6+uQ/OSlGw2okOFGx33ixAl17NhR//znP/XKK6+oZMmS+uabbzRw4EAlJyerWLFiWdbp5+fnsOzp6akNGzbo22+/tXVjfO6557Rr1y7bM2wAUJRwJwkACpGAgACVK1dO27dvdyjfvn27atWqJUm2kczsn3vJ2Gb48OFq3769ateuLYvFoj///DNX48u4e2E/Ol9uDkW+a9cuh+WMZ6s8PT11zz33KC0tTefOnVPVqlUdptDQUJf2U7NmTe3YscMh2du+fbv8/f1Vvnx5p+upV6+eNm3a5NK+7e3bt09Wq1UzZ85U48aNVb16dZ05cyZHdZlMJjVp0kSTJk3SgQMHZDabtXbt2hzHBgC3M5IkAChk/vWvf2n69On64IMPdOTIEY0bN04HDx7UiBEjJEllypSRr6+v1q1bp7i4OF28eFGSVK1aNS1fvlw///yzdu3apd69ezt1p8MVvr6+aty4saZNm6aff/5ZW7du1fPPP59r9Z86dUrR0dE6cuSI3n//fb3++uu2465evbp69+6tPn36aM2aNTp+/Lh2796tqVOn6vPPP3dpP08//bROnz6tYcOG6fDhw/r44481YcIERUdHuzRc+Pjx47Vnzx49/fTT+v7773X48GEtWLDA6eS0atWqSklJ0euvv67ffvtNy5cv18KFC106Fik9uZwyZYr27t2rU6dOac2aNfrjjz9Us2ZNl+sCgMKAJAkACpnhw4crOjpazzzzjOrWrat169bpk08+UbVq1SRJXl5emjt3rt58802VK1dOnTt3liQtXrxY58+f17333qsnnnhCw4cPz3Jkt1v1zjvvKDU1VeHh4Ro5cqRtRLjc0KdPH129elWNGjXSkCFDNGLECA0aNMi2PiYmRn369NEzzzyjGjVqqEuXLtqzZ48qVqzo0n7uuOMO/fe//9Xu3btVv359/eMf/9DAgQNdTviqV6+uL7/8Ut99950aNWqkiIgIffzxx/Lycq43fP369TVr1ixNnz5dderU0XvvvaepU6e6FIOUfgfy66+/Vvv27VW9enU9//zzmjlzptq1a+dyXQBQGJgMg/E9AQAAACADd5IAAAAAwA5JEgAAAADYIUkCAAAAADskSQAAAABghyQJAAAAAOyQJAEAAACAHZIkAAAAALBDkgQAAAAAdkiSAAAAAMAOSRIAAAAA2CFJAgAAAAA7/wf9zJqtAQwoewAAAABJRU5ErkJggg==\n"},"metadata":{}}],"source":["plot_compare('total_chairs', 'Total number of chairs')"]},{"cell_type":"markdown","metadata":{"id":"r8Xf-gU-XvBe"},"source":["Big Mountain has amongst the highest number of total chairs, resorts with more appear to be outliers."]},{"cell_type":"markdown","metadata":{"id":"E_dsjupAXvBe"},"source":["### 5.8.5 Fast quads"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"KTR4M9XWXvBe","executionInfo":{"status":"ok","timestamp":1721138944231,"user_tz":240,"elapsed":432,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"a6602cfb-5996-447a-e395-d7f126119a7b"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('fastQuads', 'Number of fast quads')"]},{"cell_type":"markdown","metadata":{"id":"MUTsI3PeXvBe"},"source":["Most resorts have no fast quads. Big Mountain has 3, which puts it high up that league table. There are some values much higher, but they are rare."]},{"cell_type":"markdown","metadata":{"id":"m_tR6skVXvBe"},"source":["### 5.8.6 Runs"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"D6ym1RcSXvBe","executionInfo":{"status":"ok","timestamp":1721138957714,"user_tz":240,"elapsed":670,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"25c17a82-422d-4479-bc4a-5b0b89f382ef"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('Runs', 'Total number of runs')"]},{"cell_type":"markdown","metadata":{"id":"Mw5KtnKZXvBf"},"source":["Big Mountain compares well for the number of runs. There are some resorts with more, but not many."]},{"cell_type":"markdown","metadata":{"id":"vkZqqNhgXvBf"},"source":["### 5.8.7 Longest run"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"c4KrILuHXvBf","executionInfo":{"status":"ok","timestamp":1721138967787,"user_tz":240,"elapsed":422,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"cd787e98-0eb2-4bed-964e-6713ba9b3ab8"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('LongestRun_mi', 'Longest run length (miles)')"]},{"cell_type":"markdown","metadata":{"id":"iSwKop2eXvBf"},"source":["Big Mountain has one of the longest runs. Although it is just over half the length of the longest, the longer ones are rare."]},{"cell_type":"markdown","metadata":{"id":"Z8bTdjxdXvBf"},"source":["### 5.8.8 Trams"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"T-V_w4i8XvBf","executionInfo":{"status":"ok","timestamp":1721138984560,"user_tz":240,"elapsed":631,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"1ca04ab5-9a39-4c6d-c512-343e49967dcd"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["plot_compare('trams', 'Number of trams')"]},{"cell_type":"markdown","metadata":{"id":"ceKY6psBXvBf"},"source":["The vast majority of resorts, such as Big Mountain, have no trams."]},{"cell_type":"markdown","metadata":{"id":"zg1NanTuXvBg"},"source":["### 5.8.9 Skiable terrain area"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"Z3EZZGb3XvBg","executionInfo":{"status":"ok","timestamp":1721138998784,"user_tz":240,"elapsed":488,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"62c09821-c3d6-4f9f-db3a-3e92ac9ffb0d"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"iVBORw0KGgoAAAANSUhEUgAAA1IAAAHWCAYAAAB9mLjgAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABoC0lEQVR4nO3dd3gUVdvH8d+mB9JoKUgIvRcFFKM0pYQiUkWKSMdHQAUElccCiAqCUiyI5ZFiQ1FEbCgdBERAilQBQVAIPUCA1D3vH3mzsKSQgYQNy/dzXXNdszNnztwze2Z3752ZMzZjjBEAAAAAIMc8XB0AAAAAANxoSKQAAAAAwCISKQAAAACwiEQKAAAAACwikQIAAAAAi0ikAAAAAMAiEikAAAAAsIhECgAAAAAsIpECAAAAAItIpHBTsdlsGjRoULZl9u/fL5vNphkzZliuf9SoUbLZbDp+/PgVy5YqVUo9e/a0vI4b1YwZM2Sz2bR//35Xh+ISdrtd1apV08svv+zqUK5J586d1alTp2uux2azadSoUY7X17N99OzZU6VKlXK8Tj/mX3vttTxft3Txc8IV4uPj1bdvX4WHh8tms2nw4MEuieNm4I6f8enH6fr1610dSqasfAcDuYFECm7hjz/+UMeOHRUVFSU/Pz/dcsstatq0qd58801Xh3bdTJ069aqSP1wfn332mQ4ePHjFRD6/e/rpp/XVV19p8+bNrg5F58+f16hRo7Rs2TJXh5JBfo3tlVde0YwZM/Too4/qo48+Uvfu3V0dUr6wevVqjRo1SnFxca4O5abyyiuvaN68ea4OA7hqJFK44a1evVp16tTR5s2b1a9fP7311lvq27evPDw8NGXKFMv1RUVF6cKFCzfcD4z8nkh1795dFy5cUFRUlKtDcYkJEyaoc+fOCg4OdnUo1+S2225TnTp19Prrr+dqvVfTPs6fP6/Ro0dbTlbef/997dq1y2KE1mQX23PPPacLFy7k6fqzsmTJEt15550aOXKkHnroIdWuXdslceQ3q1ev1ujRo3M1kdq1a5fef//9XKvPHZFI4Ubn5eoAgGv18ssvKzg4WOvWrVNISIjTvKNHj1quz2azyc/PL5eiu7GlpKTIbrfLx8cnw7xz586pYMGCOa7L09NTnp6euRneVbPb7UpKSrpu7/PGjRu1efPmXE8+riQhIUE+Pj7y8Mjd/8w6deqkkSNHaurUqQoICMiVOq9H+0hvs97e3nm6nivx8vKSl5drvn6PHj2qKlWq5Fp92X1GZOZ6H3tXYvVzzApfX988qdcV0j9LIBljlJCQIH9/f1eHgnyAM1K44e3du1dVq1bNkERJUmho6BWXf+mll+Th4eG4DDCze6S2bNminj17qkyZMvLz81N4eLh69+6tEydOZFrn8ePH1alTJwUFBalIkSJ64oknlJCQcMVY4uLiNHjwYEVGRsrX11flypXTq6++Krvdnu1ypUqV0rZt27R8+XLZbDbZbDY1atTIUr2X3icyefJklS1bVr6+vtq+fbvjuvPt27era9euKlSokOrVq2dp32R2D0ypUqV033336ZdfftEdd9whPz8/lSlTRrNmzbrivpKk1157TXfddZeKFCkif39/1a5dW19++WWGcun3xn3yySeqWrWqfH19tWDBAknSv//+q969eyssLEy+vr6qWrWqPvzwQ6flk5KS9MILL6h27doKDg5WwYIFVb9+fS1dujRHcc6bN08+Pj5q0KCB0/S///5bAwYMUMWKFeXv768iRYrogQceyPQ+obi4OA0ZMkSlSpWSr6+vSpQooYcffthxL8CyZctks9k0e/ZsPffcc7rllltUoEABnTlzRpK0du1aNW/eXMHBwSpQoIAaNmyoVatWOa3j7NmzGjx4sGMdoaGhatq0qX7//Xenck2bNtW5c+e0cOHCK257YmKihgwZomLFiikwMFD333+//vnnnwzlMmsf69evV0xMjIoWLSp/f3+VLl1avXv3lpTWXosVKyZJGj16tKPdp9931bNnTwUEBGjv3r1q2bKlAgMD1a1bN8e8S++RutSkSZMUFRUlf39/NWzYUFu3bnWa36hRI6djK92ldV4ptszukUpJSdGYMWMcx12pUqX03//+V4mJiU7lrvaYSW8f+/bt0/fff++IKX1/Hz16VH369FFYWJj8/PxUs2ZNzZw506mO7D4jsnKtx54kvfnmm6pataoKFCigQoUKqU6dOvr000+dymzcuFEtWrRQUFCQAgIC1LhxY/36669OZdLb2PLlyzVgwACFhoaqRIkSGjVqlIYPHy5JKl26dIZ9s3DhQtWrV08hISEKCAhQxYoV9d///jfb/S1lvEcqff2rVq3S0KFDVaxYMRUsWFDt2rXTsWPHrlhfeps+cOCA7rvvPgUEBOiWW27R22+/LSntEvd7771XBQsWVFRUVIZ9dPLkSQ0bNkzVq1dXQECAgoKC1KJFiwyX6V7ps+Ryp06d0h133KESJUo4zvQmJiZq5MiRKleunHx9fRUZGamnnnrKqT3bbDadO3dOM2fOdOzzK91TlpO2IKV9Xvbs2VMhISEKDg5Wr169dP78eacy06dP17333qvQ0FD5+vqqSpUqeueddzLUlX7M/fTTT6pTp478/f317rvvOtZzNd/ZcB+ckcINLyoqSmvWrNHWrVtVrVo1S8s+99xzeuWVV/Tuu++qX79+WZZbuHCh/vrrL/Xq1Uvh4eHatm2b3nvvPW3btk2//vprhh9FnTp1UqlSpTR27Fj9+uuveuONN3Tq1Klsf+ycP39eDRs21L///qtHHnlEJUuW1OrVqzVixAgdPnxYkydPznLZyZMn67HHHlNAQICeffZZSVJYWNhV1Tt9+nQlJCSof//+8vX1VeHChR3zHnjgAZUvX16vvPKKjDFXtW8ut2fPHnXs2FF9+vRRjx499OGHH6pnz56qXbu2qlatmu2yU6ZM0f33369u3bopKSlJs2fP1gMPPKDvvvtOrVq1ciq7ZMkSffHFFxo0aJCKFi2qUqVK6ciRI7rzzjsdP/aKFSumH3/8UX369NGZM2ccN+KfOXNGH3zwgbp06aJ+/frp7Nmz+t///qeYmBj99ttvuvXWW7ONc/Xq1apWrVqGMyHr1q3T6tWr1blzZ5UoUUL79+/XO++8o0aNGmn79u0qUKCApLQOAurXr68dO3aod+/eqlWrlo4fP6758+frn3/+UdGiRR11jhkzRj4+Pho2bJgSExPl4+OjJUuWqEWLFqpdu7ZGjhwpDw8Px4+IlStX6o477pAk/ec//9GXX36pQYMGqUqVKjpx4oR++eUX7dixQ7Vq1XKso0qVKvL399eqVavUrl27bLe9b9+++vjjj9W1a1fdddddWrJkSYb3JjNHjx5Vs2bNVKxYMT3zzDMKCQnR/v37NXfuXElSsWLF9M477+jRRx9Vu3bt1L59e0lSjRo1HHWkpKQoJiZG9erV02uvvebYn1mZNWuWzp49q4EDByohIUFTpkzRvffeqz/++MNxPOVETmK7XN++fTVz5kx17NhRTz75pNauXauxY8dqx44d+vrrr53KXs0xU7lyZX300UcaMmSISpQooSeffNIR64ULF9SoUSPt2bNHgwYNUunSpTVnzhz17NlTcXFxeuKJJ5zqyu4zIjPXcuy9//77evzxx9WxY0fHH1JbtmzR2rVr1bVrV0nStm3bVL9+fQUFBempp56St7e33n33XTVq1EjLly9X3bp1neIZMGCAihUrphdeeEHnzp1TixYt9Oeff+qzzz7TpEmTHMdTsWLFtG3bNt13332qUaOGXnzxRfn6+mrPnj0Z/oSw4rHHHlOhQoU0cuRI7d+/X5MnT9agQYP0+eefX3HZ1NRUtWjRQg0aNND48eP1ySefaNCgQSpYsKCeffZZdevWTe3bt9e0adP08MMPKzo6WqVLl5Yk/fXXX5o3b54eeOABlS5dWkeOHNG7776rhg0bavv27SpevLjTujL7LLnc8ePH1bRpU508eVLLly9X2bJlZbfbdf/99+uXX35R//79VblyZf3xxx+aNGmS/vzzT8elfB999JH69u2rO+64Q/3795cklS1bNsttz0lbSNepUyeVLl1aY8eO1e+//64PPvhAoaGhevXVVx1l3nnnHVWtWlX333+/vLy89O2332rAgAGy2+0aOHCgU327du1Sly5d9Mgjj6hfv36qWLHiNX1nw40Y4Ab3888/G09PT+Pp6Wmio6PNU089ZX766SeTlJSUoawkM3DgQGOMMU8++aTx8PAwM2bMcCqzb98+I8lMnz7dMe38+fMZ6vrss8+MJLNixQrHtJEjRxpJ5v7773cqO2DAACPJbN682TEtKirK9OjRw/F6zJgxpmDBgubPP/90WvaZZ54xnp6e5sCBA9nuh6pVq5qGDRtmmJ7TetO3OygoyBw9etSpbPp2denSJUP9Od0306dPN5LMvn37HNOioqIylDt69Kjx9fU1Tz75ZLbbm9m6k5KSTLVq1cy9997rNF2S8fDwMNu2bXOa3qdPHxMREWGOHz/uNL1z584mODjYUX9KSopJTEx0KnPq1CkTFhZmevfufcU4S5QoYTp06HDF+I0xZs2aNUaSmTVrlmPaCy+8YCSZuXPnZihvt9uNMcYsXbrUSDJlypRxqtdut5vy5cubmJgYR9n0dZcuXdo0bdrUMS04ONhxfFxJhQoVTIsWLbIts2nTJiPJDBgwwGl6165djSQzcuRIx7TL28fXX39tJJl169ZlWf+xY8cy1JOuR48eRpJ55plnMp0XFRXleJ3e9v39/c0///zjmL527VojyQwZMsQxrWHDhpkeZ5fXmV1s6cdTuvT91LdvX6dyw4YNM5LMkiVLHNOu9ZiJiooyrVq1cpo2efJkI8l8/PHHjmlJSUkmOjraBAQEmDNnzhhjsv+MyMq1Hntt2rQxVatWzXYdbdu2NT4+Pmbv3r2OaYcOHTKBgYGmQYMGjmnpbaxevXomJSXFqY4JEyZk+HwyxphJkyYZSebYsWM52t5LXf4Zn77+Jk2aOB2LQ4YMMZ6eniYuLi7b+tLb9CuvvOKYdurUKePv729sNpuZPXu2Y/rOnTsztL+EhASTmprqVOe+ffuMr6+vefHFFx3TsvosuXQb1q1bZw4fPmyqVq1qypQpY/bv3+8o89FHHxkPDw+zcuVKp2WnTZtmJJlVq1Y5phUsWNBpH2UnJ20h/di6/HO5Xbt2pkiRIk7TMvv8jYmJMWXKlHGaln7MLViwwGn6tX5nwz1waR9ueE2bNtWaNWt0//33a/PmzRo/frxiYmJ0yy23aP78+RnKG2M0aNAgTZkyRR9//LF69OhxxXVcei10QkKCjh8/rjvvvFOSMlz2JCnDv1mPPfaYJOmHH37Ich1z5sxR/fr1VahQIR0/ftwxNGnSRKmpqVqxYsUV48yNejt06OC4LOly//nPfzJMs7pvLlelShXVr1/f8bpYsWKqWLGi/vrrrysue+m6T506pdOnT6t+/fqZrrdhw4ZO94YYY/TVV1+pdevWMsY47ZuYmBidPn3aUY+np6fj31i73a6TJ08qJSVFderUydE2njhxQoUKFco2/uTkZJ04cULlypVTSEiIU71fffWVatasmenZn8vP+PXo0cOp3k2bNmn37t3q2rWrTpw44djGc+fOqXHjxlqxYoXjMpSQkBCtXbtWhw4duuI2pben7KS398cff9xpek663E6/VPe7775TcnLyFctn5dFHH81x2bZt2+qWW25xvL7jjjtUt27dbI/b3JBe/9ChQ52mp581+v77752mX8sxk9X6w8PD1aVLF8c0b29vPf7444qPj9fy5cudymf3GZGZazn2QkJC9M8//2jdunWZ1p2amqqff/5Zbdu2VZkyZRzTIyIi1LVrV/3yyy8ZLknr169fju/HS2+H33zzTa5drtW/f3+n47Z+/fpKTU3V33//naPl+/bt6xRfxYoVVbBgQafHElSsWFEhISFObcLX19dxv2RqaqpOnDjhuFQxs8+xyz9LLvXPP/+oYcOGSk5O1ooVK5w6iZkzZ44qV66sSpUqOb239957ryTl+JLoy12pLVzq8u+q+vXr68SJE05t4dJtO336tI4fP66GDRvqr7/+0unTp52WL126tGJiYpym5dV3Nm4sXNoHt3D77bdr7ty5SkpK0ubNm/X1119r0qRJ6tixozZt2uT0JT5r1izFx8frnXfecfrhkJ2TJ09q9OjRmj17doYOLC7/wJWk8uXLO70uW7asPDw8sn1Gzu7du7Vly5Ysf6BcTccZV1Nv+mUgmclsntV9c7mSJUtmmFaoUCGdOnXqist+9913eumll7Rp06YM195fKfZjx44pLi5O7733nt57771M6790e2bOnKnXX39dO3fudPphn93+upT5/0shL3XhwgWNHTtW06dP17///utU5tJ9t3fvXnXo0CFH67k8nt27d0tStn8YnD59WoUKFdL48ePVo0cPRUZGqnbt2mrZsqUefvhhpx+ol27PlS7b/Pvvv+Xh4ZHhcp2KFStecTsaNmyoDh06aPTo0Zo0aZIaNWqktm3bqmvXrjm+id/Ly0slSpTIUVkp43ErSRUqVNAXX3yR4zquRvp+KleunNP08PBwhYSEZPiBfS3HTFbrL1++fIZOSSpXruyYf6mctvmsyls59p5++mktWrRId9xxh8qVK6dmzZqpa9euuvvuux11nT9/PtM2VblyZdntdh08eNDpkkcr8T/44IP64IMP1LdvXz3zzDNq3Lix2rdvr44dO151Jy6Xv3/pf7Lk5P3z8/PL8FkeHBysEiVKZDgeg4ODneq02+2aMmWKpk6dqn379ik1NdUxr0iRIhnWld1+6t69u7y8vLRjxw6Fh4c7zdu9e7d27NiR699lV2oLl8puHwcFBUmSVq1apZEjR2rNmjUZ7p86ffq0Uw+rme2LvPrOxo2FRApuxcfHR7fffrtuv/12VahQQb169dKcOXM0cuRIR5m7775bmzZt0ltvvaVOnTpd8fp+Ke1669WrV2v48OG69dZbFRAQILvdrubNm+foX8qcPHzTbreradOmeuqppzKdX6FChSvWkRv1ZtcTUWbzrnXfZPXPcGaJx6VWrlyp+++/Xw0aNNDUqVMVEREhb29vTZ8+PdObjy+PPT22hx56KMskI/2elo8//lg9e/ZU27ZtNXz4cIWGhsrT01Njx47V3r17r7iNRYoUyfRH0mOPPabp06dr8ODBio6OVnBwsGw2mzp37nzV/35ntZ0TJkzI8l6u9J73OnXqpPr16+vrr7/Wzz//rAkTJujVV1/V3Llz1aJFC6dlTp06lWnikVtsNpu+/PJL/frrr/r222/1008/qXfv3nr99df166+/5qi3wEv/gc/NuDJrm5f+KL2WunPiao+Z3GK1t7JrOfYqV66sXbt26bvvvtOCBQv01VdfaerUqXrhhRc0evToq4jeWvz+/v5asWKFli5dqu+//14LFizQ559/rnvvvVc///zzVfU0eS3vX1bL5qTOV155Rc8//7x69+6tMWPGqHDhwvLw8NDgwYMz/bzJbj+1b99es2bN0pQpUzR27FineXa7XdWrV9fEiRMzXTYyMjLLerNjpS1caX/s3btXjRs3VqVKlTRx4kRFRkbKx8dHP/zwgyZNmpRhf2S2L/LqOxs3FhIpuK06depIkg4fPuw0vVy5cho/frwaNWqk5s2ba/HixQoMDMyynlOnTmnx4sUaPXq0XnjhBcf09H/6M7N7926nf7D27Nkju92eZU9hUtpZq/j4eDVp0uRKm5aprH6EXWu92bmafZNbvvrqK/n5+emnn35yOkMxffr0HC2f3otcamrqFffNl19+qTJlymju3LlO+/nSBD07lSpV0r59+zKtt0ePHk7doickJGR4lk3ZsmUz9B6XU+lng4KCgnLUBiIiIjRgwAANGDBAR48eVa1atfTyyy87JVIpKSk6ePCg7r///mzrioqKkt1u1969e53OGFh5htOdd96pO++8Uy+//LI+/fRTdevWTbNnz1bfvn1znHjkVGbt9s8//3Q6bgsVKpTpJXSXn7WxElv6ftq9e7fjLJAkHTlyRHFxcXn+7LWoqCht2bJFdrvdKfHcuXOnY35usnLsSVLBggX14IMP6sEHH1RSUpLat2+vl19+WSNGjFCxYsVUoECBTNvUzp075eHhkaMf7tm9Xx4eHmrcuLEaN26siRMn6pVXXtGzzz6rpUuX5snnal758ssvdc899+h///uf0/S4uDinDmty4rHHHlO5cuX0wgsvKDg4WM8884xjXtmyZbV582Y1btz4iseB1WM4u7ZgpUv9b7/9VomJiZo/f77T2Ssrlx3m5XcrbhzcI4Ub3tKlSzP9Jy/9voPMLvmoUaOGfvjhB+3YsUOtW7fO9uGY6f9sXb6O7HrkSe+ONl161+qX/6t/qU6dOmnNmjX66aefMsyLi4tTSkpKlstKaV8wmT1M8lrrzc7V7Jvc4unpKZvN5nQmYP/+/Tl+uKOnp6c6dOigr776KtMk5dLuiDPbzrVr12rNmjU5Wld0dLS2bt2aoStrT0/PDPvuzTffzHB2o0OHDo5LVi93pX+xa9eurbJly+q1115TfHx8hvnp25mamprhUszQ0FAVL148Q9zbt29XQkKC7rrrrmzXnd7e33jjDafpOWkfp06dyrBt6WfU0uNJ74Uvtx6iOm/ePP3777+O17/99pvWrl3rdNyWLVtWO3fudGofmzdvztCLm5XYWrZsKSnjfkn/Rz8nvRxei5YtWyo2Ntap17iUlBS9+eabCggIUMOGDXN1fVaOvcsfo+Dj46MqVarIGKPk5GR5enqqWbNm+uabb5wunT5y5Ig+/fRT1atXz3EpV3bSnyV1+ft18uTJDGUvb4c3isw+b+bMmePU5q14/vnnNWzYMI0YMcKp2/BOnTrp33//zfRhxBcuXNC5c+ccr7P63srMldqCFZl9pp8+fTrHf8RJefvdihsHZ6Rww3vsscd0/vx5tWvXTpUqVVJSUpJWr16tzz//XKVKlVKvXr0yXe7OO+/UN998o5YtW6pjx46aN29epg/qDAoKcnQ1m5ycrFtuuUU///xzpmcY0u3bt0/333+/mjdvrjVr1ji6f65Zs2aWywwfPlzz58/Xfffd5+jK+Ny5c/rjjz/05Zdfav/+/dn+a1i7dm298847eumll1SuXDmFhobq3nvvveZ6s3M1+ya3tGrVShMnTlTz5s3VtWtXHT16VG+//bbKlSunLVu25KiOcePGaenSpapbt6769eunKlWq6OTJk/r999+1aNEix4+o++67T3PnzlW7du3UqlUr7du3T9OmTVOVKlUyTU4u16ZNG40ZM0bLly9Xs2bNHNPvu+8+ffTRRwoODlaVKlW0Zs0aLVq0KMP9CsOHD9eXX36pBx54QL1791bt2rV18uRJzZ8/X9OmTcu2XXl4eOiDDz5QixYtVLVqVfXq1Uu33HKL/v33Xy1dulRBQUH69ttvdfbsWZUoUUIdO3ZUzZo1FRAQoEWLFmndunUZHiS8cOFCFShQQE2bNs12u2+99VZ16dJFU6dO1enTp3XXXXdp8eLF2rNnzxX32cyZMzV16lS1a9dOZcuW1dmzZ/X+++8rKCjIkXj4+/urSpUq+vzzz1WhQgUVLlxY1apVs/wYhHTlypVTvXr19OijjyoxMVGTJ09WkSJFnC7d6d27tyZOnKiYmBj16dNHR48e1bRp01S1atUMN7LnNLaaNWuqR48eeu+99xQXF6eGDRvqt99+08yZM9W2bVvdc889V7U9OdW/f3+9++676tmzpzZs2KBSpUrpyy+/1KpVqzR58uRsz9hfrZwee82aNVN4eLjuvvtuhYWFaceOHXrrrbfUqlUrR1wvvfSS41lPAwYMkJeXl959910lJiZq/PjxOYqndu3akqRnn31WnTt3lre3t1q3bq0XX3xRK1asUKtWrRQVFaWjR49q6tSpKlGihONZejeK++67Ty+++KJ69eqlu+66S3/88Yc++eSTTO+BzKkJEybo9OnTGjhwoAIDA/XQQw+pe/fu+uKLL/Sf//xHS5cu1d13363U1FTt3LlTX3zxheN5TFLafl+0aJEmTpyo4sWLq3Tp0hm6q0+Xk7aQU82aNZOPj49at26tRx55RPHx8Xr//fcVGhqa4SqWrOTldytuINetf0Agj/z444+md+/eplKlSiYgIMD4+PiYcuXKmccee8wcOXLEqawu6f483TfffGO8vLzMgw8+aFJTUzPt/vyff/4x7dq1MyEhISY4ONg88MAD5tChQxm6l03venX79u2mY8eOJjAw0BQqVMgMGjTIXLhwwWm9l3eNa4wxZ8+eNSNGjDDlypUzPj4+pmjRouauu+4yr732WqbduV8qNjbWtGrVygQGBhpJTl0056Te9O2eMGFChrrTtyuzLoBzum+y6v788q6Yjcm6i+nL/e9//zPly5c3vr6+plKlSmb69OkZupY2JvP3Pd2RI0fMwIEDTWRkpPH29jbh4eGmcePG5r333nOUsdvt5pVXXjFRUVHG19fX3Hbbbea7777L0OV1dmrUqGH69OnjNO3UqVOmV69epmjRoiYgIMDExMSYnTt3Zto2Tpw4YQYNGmRuueUW4+PjY0qUKGF69Ojh6D46vcviOXPmZLr+jRs3mvbt25siRYoYX19fExUVZTp16mQWL15sjDEmMTHRDB8+3NSsWdMEBgaaggULmpo1a5qpU6dmqKtu3brmoYceytF2X7hwwTz++OOmSJEipmDBgqZ169bm4MGDV2wfv//+u+nSpYspWbKk8fX1NaGhoea+++4z69evd6p/9erVpnbt2sbHx8epzh49epiCBQtmGlNW3Z9PmDDBvP766yYyMtL4+vqa+vXrOz2yIN3HH39sypQpY3x8fMytt95qfvrpp0zbQlaxZdZGk5OTzejRo03p0qWNt7e3iYyMNCNGjDAJCQlO5a71mMlq+SNHjjjaoo+Pj6levbrTZ+Dl+ymnrvXYe/fdd02DBg0c7bZs2bJm+PDh5vTp0051/f777yYmJsYEBASYAgUKmHvuucesXr3aqcylXXdnZsyYMeaWW24xHh4ejra4ePFi06ZNG1O8eHHj4+Njihcvbrp06ZKhy+vMZNX9+eXrTz92ly5dmm19WbXphg0bZtot+OXvdUJCgnnyySdNRESE8ff3N3fffbdZs2ZNhraT3WdJZtuQmppqunTpYry8vMy8efOMMWnd57/66qumatWqxtfX1xQqVMjUrl3bjB492um927lzp2nQoIHx9/c3krLtCj0nbSGr76rMvn/mz59vatSoYfz8/EypUqXMq6++aj788MMcf08Zc23f2XAPNmOu092pAHAT++ijjzRw4EAdOHDA0aXyjWjTpk2qVauWfv/99ys+iBgAAHdGIgUA14HdbleNGjXUpUsXPfvss64O56ql9yiY112CAwCQ35FIAQAAAIBF9NoHAAAAABaRSAEAAACARSRSAAAAAGARiRQAAAAAWMQDeZXWm9ahQ4cUGBgom83m6nAAAAAAuIgxRmfPnlXx4sXl4ZH1eScSKUmHDh1SZGSkq8MAAAAAkE8cPHhQJUqUyHI+iZSkwMBASWk7KygoyMXR4KrY7dKRI2njYWFSNv8eAAAAAFk5c+aMIiMjHTlCVkikJMflfEFBQSRSN6oLF6Ru3dLGV66U/P1dGw8AAABuaFe65Ye/7QEAAADAIhIpAAAAALCIRAoAAAAALOIeKQAAANzwjDFKSUlRamqqq0NBPufp6SkvL69rfuwRiRQAAABuaElJSTp8+LDOnz/v6lBwgyhQoIAiIiLk4+Nz1XWQSAEAAOCGZbfbtW/fPnl6eqp48eLy8fG55jMNcF/GGCUlJenYsWPat2+fypcvn+1Dd7NDIgX34OkpPfDAxXEAAHBTSEpKkt1uV2RkpAoUKODqcHAD8Pf3l7e3t/7++28lJSXJz8/vquohkYJ78PGRnn7a1VEAAAAXudqzCrg55UZ7ocUBAAAAgEWckYJ7MEaKi0sbDwmRuDYaAAAAeYgzUnAPCQlS06ZpQ0KCq6MBAADIFfv375fNZtOmTZtcHUq+UqpUKU2ePNmlMZBIAQAAAC7Qs2dP2Ww2x1CkSBE1b95cW7ZscZSJjIzU4cOHVa1atWtaV6lSpWSz2TR79uwM86pWrSqbzaYZM2Zc0zquhs1m07x58ywvt27dOvXv3z/3A7KARAoAAABwkebNm+vw4cM6fPiwFi9eLC8vL913332O+Z6engoPD5eX17XfkRMZGanp06c7Tfv1118VGxurggULXnP911OxYsVc3ksjiRQAAADc04ULWQ9JSTkvm5iYs7JXwdfXV+Hh4QoPD9ett96qZ555RgcPHtSxY8ckZX5p3/z581W+fHn5+fnpnnvu0cyZM2Wz2RSXfr94Frp166bly5fr4MGDjmkffvihunXrliFRO3DggNq0aaOAgAAFBQWpU6dOOnLkiGN+z5491bZtW6dlBg8erEaNGjleN2rUSI8//rieeuopFS5cWOHh4Ro1apRjfqlSpSRJ7dq1k81mc7zeu3ev2rRpo7CwMAUEBOj222/XokWLnNZ1+aV9NptNH3zwgdq1a6cCBQqofPnymj9/frb741qRSAEAAMA91a+f9TB8uHPZpk2zLvvYY85lW7fOvNw1io+P18cff6xy5cqpSJEimZbZt2+fOnbsqLZt22rz5s165JFH9Oyzz+ao/rCwMMXExGjmzJmSpPPnz+vzzz9X7969ncrZ7Xa1adNGJ0+e1PLly7Vw4UL99ddfevDBBy1v08yZM1WwYEGtXbtW48eP14svvqiFCxdKSrs8T5KmT5+uw4cPO17Hx8erZcuWWrx4sTZu3KjmzZurdevWOnDgQLbrGj16tDp16qQtW7aoZcuW6tatm06ePGk55pyi1758qNQz3+dZ3fvHtcqzugEAAGDNd999p4CAAEnSuXPnFBERoe+++y7L5xy9++67qlixoiZMmCBJqlixorZu3aqXX345R+vr3bu3nnzyST377LP68ssvVbZsWd16661OZRYvXqw//vhD+/btU2RkpCRp1qxZqlq1qtatW6fbb789x9tXo0YNjRw5UpJUvnx5vfXWW1q8eLGaNm2qYsWKSZJCQkIUHh7uWKZmzZqqWbOm4/WYMWP09ddfa/78+Ro0aFCW6+rZs6e6dOkiSXrllVf0xhtv6LffflPz5s1zHK8VJFIAAABwTytXZj3P09P59f+fJcnU5UnNt99efUyXueeee/TOO+9Ikk6dOqWpU6eqRYsW+u233xQVFZWh/K5duzIkMnfccUeO19eqVSs98sgjWrFihT788MMMZ6MkaceOHYqMjHQkUZJUpUoVhYSEaMeOHZYTqUtFRETo6NGj2S4THx+vUaNG6fvvv9fhw4eVkpKiCxcuXPGM1KXrKliwoIKCgq64rmtBIgX34Okppd+YefkHIwAAuDn5+7u+7BUULFhQ5cqVc7z+4IMPFBwcrPfff18vvfRSrq0nnZeXl7p3766RI0dq7dq1+vrrr6+qHg8PDxljnKYlJydnKOft7e302mazyW63Z1v3sGHDtHDhQr322msqV66c/P391bFjRyVdfl9bLqzrWpBIwT34+EiX3LwIAABwI7LZbPLw8NCFLDqvqFixon744Qenaen3FuVU79699dprr+nBBx9UoUKFMsyvXLmyDh48qIMHDzrOSm3fvl1xcXGqUqWKpLRe87Zu3eq03KZNmzIkM1fi7e2t1NRUp2mrVq1Sz5491a5dO0lpZ6j2799vqd7rgc4mAAAAABdJTExUbGysYmNjtWPHDj322GOKj49X69atMy3/yCOPaOfOnXr66af1559/6osvvnA8/8lms+VonZUrV9bx48czdIWerkmTJqpevbq6deum33//Xb/99psefvhhNWzYUHXq1JEk3XvvvVq/fr1mzZql3bt3a+TIkRkSq5woVaqUFi9erNjYWJ06dUpS2r1Uc+fO1aZNm7R582Z17do1T88sXS2XJlJjx47V7bffrsDAQIWGhqpt27batWuXU5lGjRo5PajMZrPpP//5j1OZAwcOqFWrVipQoIBCQ0M1fPhwpaSkXM9NgasZc7Hr0ctOMwMAAORXCxYsUEREhCIiIlS3bl2tW7dOc+bMcepG/FKlS5fWl19+qblz56pGjRp65513HL32+fr65ni9RYoUkX8WlyjabDZ98803KlSokBo0aKAmTZqoTJky+vzzzx1lYmJi9Pzzz+upp57S7bffrrNnz+rhhx/O+Yb/v9dff10LFy5UZGSkbrvtNknSxIkTVahQId11111q3bq1YmJiVKtWLct15zWbufzixuuoefPm6ty5s26//XalpKTov//9r7Zu3art27c7HgrWqFEjVahQQS+++KJjuQIFCigoKEiSlJqaqltvvVXh4eGaMGGCDh8+rIcfflj9+vXTK6+8kqM4zpw5o+DgYJ0+fdpRryvRa99VuHDhYrejK1fm6rXLAAAg/0pISNC+fftUunRp+fn5uTocl3j55Zc1bdo0p+dDIXvZtZuc5gYuvUdqwYIFTq9nzJih0NBQbdiwQQ0aNHBML1CggFOXiJf6+eeftX37di1atEhhYWG69dZbNWbMGD399NMaNWqUfHx88nQbAAAAgOtp6tSpuv3221WkSBGtWrVKEyZMyLZbcOSNfHWP1OnTpyVJhQsXdpr+ySefqGjRoqpWrZpGjBih8+fPO+atWbNG1atXV1hYmGNaTEyMzpw5o23btmW6nsTERJ05c8ZpAAAAAG4Eu3fvVps2bVSlShWNGTNGTz75pEbR6dZ1l2967bPb7Ro8eLDuvvtuVatWzTG9a9euioqKUvHixbVlyxY9/fTT2rVrl+bOnStJio2NdUqiJDlex8bGZrqusWPHavTo0Xm0JQAAAEDemTRpkiZNmuTqMG56+SaRGjhwoLZu3apffvnFaXr//v0d49WrV1dERIQaN26svXv3qmzZsle1rhEjRmjo0KGO12fOnHF64BgAAAAAZCdfXNo3aNAgfffdd1q6dKlKlCiRbdm6detKkvbs2SNJCg8P15EjR5zKpL/O6r4qX19fBQUFOQ0AAAC4cbmw/zTcgHKjvbg0kTLGaNCgQfr666+1ZMkSlS5d+orLbNq0SZIUEREhSYqOjtYff/yho0ePOsosXLhQQUFBjgeGAQAAwD2lPwD20nvogStJby9WHyB8KZde2jdw4EB9+umn+uabbxQYGOi4pyk4OFj+/v7au3evPv30U7Vs2VJFihTRli1bNGTIEDVo0EA1atSQJDVr1kxVqlRR9+7dNX78eMXGxuq5557TwIEDLfWljxucp6fUuPHFcQAAcFPw9PRUSEiI40/1AgUK5PjBtLj5GGN0/vx5HT16VCEhIfK8ht+NLn2OVFaNfPr06erZs6cOHjyohx56SFu3btW5c+cUGRmpdu3a6bnnnnO6HO/vv//Wo48+qmXLlqlgwYLq0aOHxo0bJy+vnOWJPEcKAADgxmWMUWxsrOLi4lwdCm4QISEhCg8PzzQfuSGeI3WlHC4yMlLLly+/Yj1RUVH64YcfcissAAAA3EBsNpsiIiIUGhqq5ORkV4eDfM7b2/uazkSlyze99gEAAADXwtPTM1d+IAM5kS967QOu2YULUp06acOFC66OBgAAAG6ORAoAAAAALCKRAgAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACwikQIAAAAAi3iOFNyDp6d0990XxwEAAIA8RCIF9+DjI02Z4uooAAAAcJPg0j4AAAAAsIhECgAAAAAsIpGCe7hwQapXL224cMHV0QAAAMDNcY8U3EdCgqsjAAAAwE2CM1IAAAAAYBGJFAAAAABYRCIFAAAAABaRSAEAAACARSRSAAAAAGARvfbBPXh4SLVqXRwHAAAA8hCJFNyDr6/03nuujgIAAAA3Cf66BwAAAACLSKQAAAAAwCISKbiHCxekJk3ShgsXXB0NAAAA3Bz3SMF9xMW5OgIAAADcJDgjBQAAAAAWkUgBAAAAgEUkUgAAAABgEYkUAAAAAFhEIgUAAAAAFtFrH9yDh4dUpcrFcQAAACAPkUjBPfj6SrNmuToKAAAA3CT46x4AAAAALCKRAgAAAACLSKTgHhISpNat04aEBFdHAwAAADfHPVJwD8ZIhw9fHAcAAADyEGekAAAAAMAiEikAAAAAsIhECgAAAAAsIpECAAAAAItIpAAAAADAInrtg3uw2aQyZS6OAwAAAHmIRAruwc9P+uILV0cBAACAmwSX9gEAAACARSRSAAAAAGARiRTcQ0KC1KlT2pCQ4OpoAAAA4Oa4RwruwRjpr78ujgMAAAB5iDNSAAAAAGARiRQAAAAAWEQiBQAAAAAWkUgBAAAAgEUkUgAAAABgEb32wT3YbFJExMVxAAAAIA+RSME9+PlJ337r6igAAABwk+DSPgAAAACwiEQKAAAAACwikYJ7SEyUHn44bUhMdHU0AAAAcHPcIwX3YLdL27dfHAcAAADyEGekAAAAAMAiEikAAAAAsMilidTYsWN1++23KzAwUKGhoWrbtq127drlVCYhIUEDBw5UkSJFFBAQoA4dOujIkSNOZQ4cOKBWrVqpQIECCg0N1fDhw5WSknI9NwUAAADATcSlidTy5cs1cOBA/frrr1q4cKGSk5PVrFkznTt3zlFmyJAh+vbbbzVnzhwtX75chw4dUvv27R3zU1NT1apVKyUlJWn16tWaOXOmZsyYoRdeeMEVmwQAAADgJmAzxhhXB5Hu2LFjCg0N1fLly9WgQQOdPn1axYoV06effqqOHTtKknbu3KnKlStrzZo1uvPOO/Xjjz/qvvvu06FDhxQWFiZJmjZtmp5++mkdO3ZMPj4+V1zvmTNnFBwcrNOnTysoKChPtzEnSj3zfZ7VvX9cqzyr26UuXJDq108bX7lS8vd3bTwAAAC4IeU0N8hX90idPn1aklS4cGFJ0oYNG5ScnKwmTZo4ylSqVEklS5bUmjVrJElr1qxR9erVHUmUJMXExOjMmTPatm1bputJTEzUmTNnnAa4gZCQtAEAAADIY/kmkbLb7Ro8eLDuvvtuVatWTZIUGxsrHx8fhVz24zgsLEyxsbGOMpcmUenz0+dlZuzYsQoODnYMkZGRubw1uO78/aVFi9IGzkYBAAAgj+WbRGrgwIHaunWrZs+enefrGjFihE6fPu0YDh48mOfrBAAAAOA+8sUDeQcNGqTvvvtOK1asUIkSJRzTw8PDlZSUpLi4OKezUkeOHFF4eLijzG+//eZUX3qvfullLufr6ytfX99c3goAAAAANwuXnpEyxmjQoEH6+uuvtWTJEpUuXdppfu3ateXt7a3Fixc7pu3atUsHDhxQdHS0JCk6Olp//PGHjh496iizcOFCBQUFqUqVKtdnQ+B6iYlS//5pQ2Kiq6MBAACAm3PpGamBAwfq008/1TfffKPAwEDHPU3BwcHy9/dXcHCw+vTpo6FDh6pw4cIKCgrSY489pujoaN15552SpGbNmqlKlSrq3r27xo8fr9jYWD333HMaOHAgZ51uJna79PvvF8cBAACAPOTSROqdd96RJDVq1Mhp+vTp09WzZ09J0qRJk+Th4aEOHTooMTFRMTExmjp1qqOsp6envvvuOz366KOKjo5WwYIF1aNHD7344ovXazMAAAAA3GRcmkjl5BFWfn5+evvtt/X2229nWSYqKko//PBDboYGAAAAAFnKN732AQAAAMCNgkQKAAAAACwikQIAAAAAi/LFc6SAXOHn5+oIAAAAcJMgkYJ78PeXfvnF1VEAAADgJsGlfQAAAABgEYkUAAAAAFhEIgX3kJQkPfFE2pCU5OpoAAAA4Oa4RwruITVVWrXq4jgAAACQhzgjBQAAAAAWkUgBAAAAgEUkUgAAAABgEYkUAAAAAFhEIgUAAAAAFpFIAQAAAIBFdH8O9+DvL61f7+ooAAAAcJPgjBQAAAAAWEQiBQAAAAAWkUjBPSQlSU8/nTYkJbk6GgAAALg5Eim4h9RUafHitCE11dXRAAAAwM2RSAEAAACARSRSAAAAAGARiRQAAAAAWEQiBQAAAAAWkUgBAAAAgEUkUgAAAABgkZerAwByhZ+ftHLlxXEAAAAgD5FIwT3YbJK/v6ujAAAAwE2CS/sAAAAAwCLOSME9JCVJr7ySNv7f/0o+Pq6NBwAAAG6NM1JwD6mp0nffpQ2pqa6OBgAAAG6ORAoAAAAALCKRAgAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACwikQIAAAAAi3iOFNyDn5+0cOHFcQAAACAPkUjBPdhsUqFCro4CAAAANwku7QMAAAAAizgjBfeQlCRNmpQ2PmSI5OPj2ngAAADg1jgjBfeQmirNmZM2pKa6OhoAAAC4ORIpAAAAALCIRAoAAAAALCKRAgAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACziOVJwD76+0vz5F8cBAACAPEQiBffg4SEVL+7qKAAAAHCT4NI+AAAAALCIM1JwD8nJ0tSpaeMDBkje3q6NBwAAAG6NM1JwDykp0kcfpQ0pKa6OBgAAAG6ORAoAAAAALCKRAgAAAACLLCdSf/31V17EAQAAAAA3DMuJVLly5XTPPffo448/VkJCQl7EBAAAAAD5muVE6vfff1eNGjU0dOhQhYeH65FHHtFvv/2WF7EBAAAAQL5kOZG69dZbNWXKFB06dEgffvihDh8+rHr16qlatWqaOHGijh07lhdxAgAAAEC+cdWdTXh5eal9+/aaM2eOXn31Ve3Zs0fDhg1TZGSkHn74YR0+fDg34wSy5+srffFF2uDr6+poAAAA4OauOpFav369BgwYoIiICE2cOFHDhg3T3r17tXDhQh06dEht2rTJzTiB7Hl4SGXKpA0edEYJAACAvGX5F+fEiRNVvXp13XXXXTp06JBmzZqlv//+Wy+99JJKly6t+vXra8aMGfr999+vWNeKFSvUunVrFS9eXDabTfPmzXOa37NnT9lsNqehefPmTmVOnjypbt26KSgoSCEhIerTp4/i4+OtbhYAAAAA5JiX1QXeeecd9e7dWz179lRERESmZUJDQ/W///3vinWdO3dONWvWVO/evdW+fftMyzRv3lzTp093vPa97LKtbt266fDhw1q4cKGSk5PVq1cv9e/fX59++qmFrcINLzlZSm8nvXpJ3t6ujQcAAABuzXIitXv37iuW8fHxUY8ePa5YrkWLFmrRokW2ZXx9fRUeHp7pvB07dmjBggVat26d6tSpI0l688031bJlS7322msqXrz4FWOAm0hJkd57L228e3cSKQAAAOQpy5f2TZ8+XXPmzMkwfc6cOZo5c2auBHWpZcuWKTQ0VBUrVtSjjz6qEydOOOatWbNGISEhjiRKkpo0aSIPDw+tXbs2yzoTExN15swZpwEAAAAAcspyIjV27FgVLVo0w/TQ0FC98soruRJUuubNm2vWrFlavHixXn31VS1fvlwtWrRQamqqJCk2NlahoaFOy3h5ealw4cKKjY3NdhuCg4MdQ2RkZK7GDQAAAMC9Wb6078CBAypdunSG6VFRUTpw4ECuBJWuc+fOjvHq1aurRo0aKlu2rJYtW6bGjRtfdb0jRozQ0KFDHa/PnDlDMgUAAAAgxyyfkQoNDdWWLVsyTN+8ebOKFCmSK0FlpUyZMipatKj27NkjSQoPD9fRo0edyqSkpOjkyZNZ3lclpd13FRQU5DQAAAAAQE5ZTqS6dOmixx9/XEuXLlVqaqpSU1O1ZMkSPfHEE05nkPLCP//8oxMnTjh6C4yOjlZcXJw2bNjgKLNkyRLZ7XbVrVs3T2MBAAAAcPOyfGnfmDFjtH//fjVu3FheXmmL2+12Pfzww5bvkYqPj3ecXZKkffv2adOmTSpcuLAKFy6s0aNHq0OHDgoPD9fevXv11FNPqVy5coqJiZEkVa5cWc2bN1e/fv00bdo0JScna9CgQercuTM99gEAAADIMzZjjLmaBf/8809t3rxZ/v7+ql69uqKioizXsWzZMt1zzz0Zpvfo0UPvvPOO2rZtq40bNyouLk7FixdXs2bNNGbMGIWFhTnKnjx5UoMGDdK3334rDw8PdejQQW+88YYCAgJyHMeZM2cUHBys06dP54vL/Eo9832e1b1/XKs8q9ul7HZp58608UqVJA/LJ1sBAACAHOcGV51IuRMSKQAAAABSznMDy5f2paamasaMGVq8eLGOHj0qu93uNH/JkiXWowUAAACAG4jlROqJJ57QjBkz1KpVK1WrVk02my0v4gKsSU6WPvssbbxLF8nb27XxAAAAwK1ZTqRmz56tL774Qi1btsyLeICrk5IivfFG2vgDD5BIAQAAIE9ZviPfx8dH5cqVy4tYAAAAAOCGYDmRevLJJzVlyhTRRwUAAACAm5XlS/t++eUXLV26VD/++KOqVq0q78suoZo7d26uBQcAAAAA+ZHlRCokJETt2rXLi1gAAAAA4IZgOZGaPn16XsQBAAAAADcMy/dISVJKSooWLVqkd999V2fPnpUkHTp0SPHx8bkaHAAAAADkR5bPSP39999q3ry5Dhw4oMTERDVt2lSBgYF69dVXlZiYqGnTpuVFnED2fH2ld9+9OA4AAADkIctnpJ544gnVqVNHp06dkr+/v2N6u3bttHjx4lwNDsgxDw+pdu20weOqTrQCAAAAOWb5jNTKlSu1evVq+fj4OE0vVaqU/v3331wLDAAAAADyK8uJlN1uV2pqaobp//zzjwIDA3MlKMCylBQpvev99u0lL8tNGwAAAMgxy9dANWvWTJMnT3a8ttlsio+P18iRI9WyZcvcjA3IueRkafz4tCE52dXRAAAAwM1Z/tv+9ddfV0xMjKpUqaKEhAR17dpVu3fvVtGiRfXZZ5/lRYwAAAAAkK9YTqRKlCihzZs3a/bs2dqyZYvi4+PVp08fdevWzanzCQAAAABwV1d1I4mXl5ceeuih3I4FAAAAAG4IlhOpWbNmZTv/4YcfvupgAAAAAOBGYDmReuKJJ5xeJycn6/z58/Lx8VGBAgVIpAAAAAC4Pcu99p06dcppiI+P165du1SvXj06mwAAAABwU8iVh+2UL19e48aN00MPPaSdO3fmRpWANT4+Unq3/Jc9LBoAAADIbbn21FIvLy8dOnQot6oDrPH0lOrVc3UUAAAAuElYTqTmz5/v9NoYo8OHD+utt97S3XffnWuBAQAAAEB+ZTmRatu2rdNrm82mYsWK6d5779Xrr7+eW3EB1qSkSD/+mDbeooXklWsnWwEAAIAMLP/atNvteREHcG2Sk6XRo9PGmzQhkQIAAECestxrHwAAAADc7Cz/bT906NAcl504caLV6gEAAAAg37OcSG3cuFEbN25UcnKyKlasKEn6888/5enpqVq1ajnK2Wy23IsSAAAAAPIRy4lU69atFRgYqJkzZ6pQoUKS0h7S26tXL9WvX19PPvlkrgcJAAAAAPmJ5XukXn/9dY0dO9aRRElSoUKF9NJLL9FrHwAAAICbguVE6syZMzp27FiG6ceOHdPZs2dzJSgAAAAAyM8sX9rXrl079erVS6+//rruuOMOSdLatWs1fPhwtW/fPtcDBHLEx0caN+7iOAAAAJCHLCdS06ZN07Bhw9S1a1clJyenVeLlpT59+mjChAm5HiCQI56eac+PAgAAAK4Dy4lUgQIFNHXqVE2YMEF79+6VJJUtW1YFCxbM9eAAAAAAID+66gfyHj58WIcPH1b58uVVsGBBGWNyMy7AmtRUadGitCE11dXRAAAAwM1ZPiN14sQJderUSUuXLpXNZtPu3btVpkwZ9enTR4UKFaLnPrhGUpL0zDNp4ytXSv7+ro0HAAAAbs3yGakhQ4bI29tbBw4cUIECBRzTH3zwQS1YsCBXgwMAAACA/MjyGamff/5ZP/30k0qUKOE0vXz58vr7779zLTAAAAAAyK8sn5E6d+6c05modCdPnpSvr2+uBAUAAAAA+ZnlRKp+/fqaNWuW47XNZpPdbtf48eN1zz335GpwAAAAAJAfWb60b/z48WrcuLHWr1+vpKQkPfXUU9q2bZtOnjypVatW5UWMAAAAAJCvWD4jVa1aNf3555+qV6+e2rRpo3Pnzql9+/bauHGjypYtmxcxAgAAAEC+YumMVHJyspo3b65p06bp2WefzauYAOu8vaWRIy+OAwAAAHnIUiLl7e2tLVu25FUswNXz8pJat3Z1FAAAALhJWL6076GHHtL//ve/vIgFAAAAAG4IljubSElJ0YcffqhFixapdu3aKliwoNP8iRMn5lpwQI6lpkpr1qSNR0dLnp6ujQcAAABuLUeJ1JYtW1StWjV5eHho69atqlWrliTpzz//dCpns9lyP0IgJ5KSpMGD08ZXrpT8/V0aDgAAANxbjhKp2267TYcPH1ZoaKj+/vtvrVu3TkWKFMnr2AAAAAAgX8rRPVIhISHat2+fJGn//v2y2+15GhQAAAAA5Gc5OiPVoUMHNWzYUBEREbLZbKpTp448s7gH5a+//srVAAEAAAAgv8lRIvXee++pffv22rNnjx5//HH169dPgYGBeR0bAAAAAORLOe61r3nz5pKkDRs26IknniCRAgAAAHDTstz9+fTp0/MiDgAAAAC4YVhOpIB8ydtbeuqpi+MAAABAHiKRgnvw8pI6dXJ1FAAAALhJ5Kj7cwAAAADARZyRgnuw26WNG9PGb7tN8uA/AgAAAOQdEim4h8RE6ZFH0sZXrpT8/V0bDwAAANwaf9sDAAAAgEUkUgAAAABgEYkUAAAAAFjk0kRqxYoVat26tYoXLy6bzaZ58+Y5zTfG6IUXXlBERIT8/f3VpEkT7d6926nMyZMn1a1bNwUFBSkkJER9+vRRfHz8ddwKAAAAADcblyZS586dU82aNfX2229nOn/8+PF64403NG3aNK1du1YFCxZUTEyMEhISHGW6deumbdu2aeHChfruu++0YsUK9e/f/3ptAgAAAICbkEt77WvRooVatGiR6TxjjCZPnqznnntObdq0kSTNmjVLYWFhmjdvnjp37qwdO3ZowYIFWrdunerUqSNJevPNN9WyZUu99tprKl68eKZ1JyYmKjEx0fH6zJkzubxlAAAAANxZvr1Hat++fYqNjVWTJk0c04KDg1W3bl2tWbNGkrRmzRqFhIQ4kihJatKkiTw8PLR27dos6x47dqyCg4MdQ2RkZN5tCK4PLy/p8cfTBi969QcAAEDeyreJVGxsrCQpLCzMaXpYWJhjXmxsrEJDQ53me3l5qXDhwo4ymRkxYoROnz7tGA4ePJjL0eO68/aWHn44bfD2dnU0AAAAcHM35V/3vr6+8vX1dXUYAAAAAG5Q+faMVHh4uCTpyJEjTtOPHDnimBceHq6jR486zU9JSdHJkycdZXCTsNul7dvTBrvd1dEAAADAzeXbRKp06dIKDw/X4sWLHdPOnDmjtWvXKjo6WpIUHR2tuLg4bdiwwVFmyZIlstvtqlu37nWPGS6UmHjx0r5LOhIBAAAA8oJLL+2Lj4/Xnj17HK/37dunTZs2qXDhwipZsqQGDx6sl156SeXLl1fp0qX1/PPPq3jx4mrbtq0kqXLlymrevLn69eunadOmKTk5WYMGDVLnzp2z7LEPAAAAAK6VSxOp9evX65577nG8Hjp0qCSpR48emjFjhp566imdO3dO/fv3V1xcnOrVq6cFCxbIz8/Pscwnn3yiQYMGqXHjxvLw8FCHDh30xhtvXPdtAQAAAHDzsBljjKuDcLUzZ84oODhYp0+fVlBQkKvDUalnvs+zuvePa5VndbvUhQtS/fpp4ytXSv7+ro0HAAAAN6Sc5gb59h4pAAAAAMivSKQAAAAAwCISKQAAAACw6KZ8IC/ckJeX1L//xXEAAAAgD/GLE+7B2/tiIgUAAADkMS7tAwAAAACLOCMF92C3S/v3p42XKiV58B8BAAAA8g6JFNxDYqLUqVPaOM+RAgAAQB7jb3sAAAAAsIhECgAAAAAsIpECAAAAAItIpAAAAADAIhIpAAAAALCIRAoAAAAALKL7c7gHLy+pe/eL4wAAAEAe4hcn3IO3t/TEE66OAgAAADcJLu0DAAAAAIs4IwX3YLdLsbFp4+Hhkgf/EQAAACDvkEjBPSQmSvffnza+cqXk7+/aeAAAAODW+NseAAAAACwikQIAAAAAi0ikAAAAAMAiEikAAAAAsIhECgAAAAAsIpECAAAAAIvo/hzuwdNTeuCBi+MAAABAHiKRgnvw8ZGeftrVUQAAAOAmwaV9AAAAAGARZ6TgHoyR4uLSxkNCJJvNldEAAADAzZFIwT0kJEhNm6aNr1wp+fu7Nh4AAAC4NS7tAwAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACwikQIAAAAAi0ikAAAAAMAiuj+He/D0lO677+I4AAAAkIdIpOAefHykUaNcHQUAAABuElzaBwAAAAAWcUYK7sEYKSEhbdzPT7LZXBsPAAAA3BpnpOAeEhKk+vXThvSECgAAAMgjJFIAAAAAYBGJFAAAAABYRCIFAAAAABaRSAEAAACARSRSAAAAAGARiRQAAAAAWMRzpOAePD2lxo0vjgMAAAB5iEQK7sHHR3r1VVdHAQAAgJsEl/YBAAAAgEUkUgAAAABgEYkU3MOFC1KdOmnDhQuujgYAAABujkQKAAAAACwikQIAAAAAi0ikAAAAAMAiEikAAAAAsIhECgAAAAAsIpECAAAAAIu8XB0AkCs8PaW77744DgAAAOQhEim4Bx8facoUV0cBAACAm0S+vrRv1KhRstlsTkOlSpUc8xMSEjRw4EAVKVJEAQEB6tChg44cOeLCiAEAAADcDPJ1IiVJVatW1eHDhx3DL7/84pg3ZMgQffvtt5ozZ46WL1+uQ4cOqX379i6MFgAAAMDNIN9f2ufl5aXw8PAM00+fPq3//e9/+vTTT3XvvfdKkqZPn67KlSvr119/1Z133nm9Q4UrXbggNW2aNr5woeTv79p4AAAA4Nby/Rmp3bt3q3jx4ipTpoy6deumAwcOSJI2bNig5ORkNWnSxFG2UqVKKlmypNasWZNtnYmJiTpz5ozTADeQkJA2AAAAAHksXydSdevW1YwZM7RgwQK988472rdvn+rXr6+zZ88qNjZWPj4+CgkJcVomLCxMsbGx2dY7duxYBQcHO4bIyMg83AoAAAAA7iZfX9rXokULx3iNGjVUt25dRUVF6YsvvpD/NVy6NWLECA0dOtTx+syZMyRTAAAAAHIsX5+RulxISIgqVKigPXv2KDw8XElJSYqLi3Mqc+TIkUzvqbqUr6+vgoKCnAYAAAAAyKkbKpGKj4/X3r17FRERodq1a8vb21uLFy92zN+1a5cOHDig6OhoF0YJAAAAwN3l60v7hg0bptatWysqKkqHDh3SyJEj5enpqS5duig4OFh9+vTR0KFDVbhwYQUFBemxxx5TdHQ0PfYBAAAAyFP5OpH6559/1KVLF504cULFihVTvXr19Ouvv6pYsWKSpEmTJsnDw0MdOnRQYmKiYmJiNHXqVBdHDZfw8JBq1bo4DgAAAOQhmzHGuDoIVztz5oyCg4N1+vTpfHG/VKlnvs+zuvePa5VndQMAAAA3upzmBvx1DwAAAAAWkUgBAAAAgEUkUnAPFy5ITZqkDRcuuDoaAAAAuLl83dkEYMllzxQDAAAA8gpnpAAAAADAIhIpAAAAALCIRAoAAAAALCKRAgAAAACLSKQAAAAAwCJ67YN78PCQqlS5OA4AAADkIRIpuAdfX2nWLFdHAQAAgJsEf90DAAAAgEUkUgAAAABgEYkU3ENCgtS6ddqQkODqaAAAAODmuEcK7sEY6fDhi+MAAABAHuKMFAAAAABYRCIFAAAAABaRSAEAAACARSRSAAAAAGARnU3cZEo9832e1r9/XKs8rR8AAADID0ik4B5sNqlMmYvjAAAAQB4ikYJ78POTvvjC1VEAAADgJsE9UgAAAABgEYkUAAAAAFhEIgX3kJAgdeqUNiQkuDoaAAAAuDnukYJ7MEb666+L4wAAAEAe4owUAAAAAFhEIgUAAAAAFpFIAQAAAIBFJFIAAAAAYBGJFAAAAABYRK99cA82mxQRcXEcAAAAyEMkUnAPfn7St9+6OgoAAADcJEikkKtKPfN9ntW9f1yrPKsbAAAAsIJ7pAAAAADAIhIpuIfEROnhh9OGxERXRwMAAAA3x6V9cA92u7R9+8VxAAAAIA9xRgoAAAAALCKRAgAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACyi1z64j5AQV0cAAACAmwSJFNyDv7+0aJGrowAAAMBNgkv7AAAAAMAiEikAAAAAsIhECu4hMVHq3z9tSEx0dTQAAABwc9wjBfdgt0u//35xHAAAAMhDnJECAAAAAItIpAAAAADAIhIpAAAAALCIRAoAAAAALKKzCdwwSj3zfZbzfJMTNeff05KkB577UYnevpbr3z+u1VXHBgAAgJsLiRTcRqKnj6tDAAAAwE2CRApuIdHbVw88NN7VYQAAAOAmwT1SAAAAAGARiRQAAAAAWMSlfXAL3qnJ+u/SDyVJr9zTW8me3i6OCAAAAO6MRApuwcNuV+1/dzjG5enigAAAAODWuLQPAAAAACwikQIAAAAAi9wmkXr77bdVqlQp+fn5qW7duvrtt99cHRIAAAAAN+UW90h9/vnnGjp0qKZNm6a6detq8uTJiomJ0a5duxQaGurq8ACVeuZ7V4dw1faPa+XqEAAAAPIdt0ikJk6cqH79+qlXr16SpGnTpun777/Xhx9+qGeeecbF0eFGcSMnOzeyG3m/k2QCAG4mefmdfSN+p97wiVRSUpI2bNigESNGOKZ5eHioSZMmWrNmTabLJCYmKjEx0fH69OnTkqQzZ87kbbA5ZE887+oQbjipyYmKt9vTxhPPy25PdXFE7iOvj4sbub3nl88MAACuh7z8zs5P36npsRhjsi13wydSx48fV2pqqsLCwpymh4WFaefOnZkuM3bsWI0ePTrD9MjIyDyJEdfH3ekjb3d3ZRhuJ3iyqyPIv9g3AADkjvz4nXr27FkFBwdnOf+GT6SuxogRIzR06FDHa7vdrpMnT6pIkSKy2WwujCwtA46MjNTBgwcVFBTk0ljgWrQFpKMt4FK0B6SjLeBStIfcY4zR2bNnVbx48WzL3fCJVNGiReXp6akjR444TT9y5IjCw8MzXcbX11e+vr5O00JCQvIqxKsSFBTEQQBJtAVcRFvApWgPSEdbwKVoD7kjuzNR6W747s99fHxUu3ZtLV682DHNbrdr8eLFio6OdmFkAAAAANzVDX9GSpKGDh2qHj16qE6dOrrjjjs0efJknTt3ztGLHwAAAADkJrdIpB588EEdO3ZML7zwgmJjY3XrrbdqwYIFGTqguBH4+vpq5MiRGS49xM2HtoB0tAVcivaAdLQFXIr2cP3ZzJX69QMAAAAAOLnh75ECAAAAgOuNRAoAAAAALCKRAgAAAACLSKQAAAAAwCISqXzk7bffVqlSpeTn56e6devqt99+c3VIuEajRo2SzWZzGipVquSYn5CQoIEDB6pIkSIKCAhQhw4dMjxc+sCBA2rVqpUKFCig0NBQDR8+XCkpKU5lli1bplq1asnX11flypXTjBkzrsfmIRsrVqxQ69atVbx4cdlsNs2bN89pvjFGL7zwgiIiIuTv768mTZpo9+7dTmVOnjypbt26KSgoSCEhIerTp4/i4+OdymzZskX169eXn5+fIiMjNX78+AyxzJkzR5UqVZKfn5+qV6+uH374Ide3F1m7Ulvo2bNnhs+J5s2bO5WhLbiHsWPH6vbbb1dgYKBCQ0PVtm1b7dq1y6nM9fxe4HeHa+WkPTRq1CjD58N//vMfpzK0BxcyyBdmz55tfHx8zIcffmi2bdtm+vXrZ0JCQsyRI0dcHRquwciRI03VqlXN4cOHHcOxY8cc8//zn/+YyMhIs3jxYrN+/Xpz5513mrvuussxPyUlxVSrVs00adLEbNy40fzwww+maNGiZsSIEY4yf/31lylQoIAZOnSo2b59u3nzzTeNp6enWbBgwXXdVjj74YcfzLPPPmvmzp1rJJmvv/7aaf64ceNMcHCwmTdvntm8ebO5//77TenSpc2FCxccZZo3b25q1qxpfv31V7Ny5UpTrlw506VLF8f806dPm7CwMNOtWzezdetW89lnnxl/f3/z7rvvOsqsWrXKeHp6mvHjx5vt27eb5557znh7e5s//vgjz/cB0lypLfTo0cM0b97c6XPi5MmTTmVoC+4hJibGTJ8+3WzdutVs2rTJtGzZ0pQsWdLEx8c7ylyv7wV+d7heTtpDw4YNTb9+/Zw+H06fPu2YT3twLRKpfOKOO+4wAwcOdLxOTU01xYsXN2PHjnVhVLhWI0eONDVr1sx0XlxcnPH29jZz5sxxTNuxY4eRZNasWWOMSfsB5uHhYWJjYx1l3nnnHRMUFGQSExONMcY89dRTpmrVqk51P/jggyYmJiaXtwZX6/Ifz3a73YSHh5sJEyY4psXFxRlfX1/z2WefGWOM2b59u5Fk1q1b5yjz448/GpvNZv79919jjDFTp041hQoVcrQFY4x5+umnTcWKFR2vO3XqZFq1auUUT926dc0jjzySq9uInMkqkWrTpk2Wy9AW3NfRo0eNJLN8+XJjzPX9XuB3R/5zeXswJi2ReuKJJ7JchvbgWlzalw8kJSVpw4YNatKkiWOah4eHmjRpojVr1rgwMuSG3bt3q3jx4ipTpoy6deumAwcOSJI2bNig5ORkp/e9UqVKKlmypON9X7NmjapXr+70cOmYmBidOXNG27Ztc5S5tI70MrSd/Gvfvn2KjY11et+Cg4NVt25dp/c+JCREderUcZRp0qSJPDw8tHbtWkeZBg0ayMfHx1EmJiZGu3bt0qlTpxxlaB/537JlyxQaGqqKFSvq0Ucf1YkTJxzzaAvu6/Tp05KkwoULS7p+3wv87sifLm8P6T755BMVLVpU1apV04gRI3T+/HnHPNqDa3m5OgBIx48fV2pqqtNBIElhYWHauXOni6JCbqhbt65mzJihihUr6vDhwxo9erTq16+vrVu3KjY2Vj4+PgoJCXFaJiwsTLGxsZKk2NjYTNtF+rzsypw5c0YXLlyQv79/Hm0drlb6e5fZ+3bp+xoaGuo038vLS4ULF3YqU7p06Qx1pM8rVKhQlu0jvQ64XvPmzdW+fXuVLl1ae/fu1X//+1+1aNFCa9askaenJ23BTdntdg0ePFh33323qlWrJknX7Xvh1KlT/O7IZzJrD5LUtWtXRUVFqXjx4tqyZYuefvpp7dq1S3PnzpVEe3A1EikgD7Vo0cIxXqNGDdWtW1dRUVH64osvSHAASJI6d+7sGK9evbpq1KihsmXLatmyZWrcuLELI0NeGjhwoLZu3apffvnF1aEgH8iqPfTv398xXr16dUVERKhx48bau3evypYte73DxGW4tC8fKFq0qDw9PTP0ynPkyBGFh4e7KCrkhZCQEFWoUEF79uxReHi4kpKSFBcX51Tm0vc9PDw803aRPi+7MkFBQSRr+VT6e5fdMR8eHq6jR486zU9JSdHJkydzpX3w2ZJ/lSlTRkWLFtWePXsk0Rbc0aBBg/Tdd99p6dKlKlGihGP69fpe4HdH/pJVe8hM3bp1Jcnp84H24DokUvmAj4+PateurcWLFzum2e12LV68WNHR0S6MDLktPj5ee/fuVUREhGrXri1vb2+n933Xrl06cOCA432Pjo7WH3/84fQjauHChQoKClKVKlUcZS6tI70MbSf/Kl26tMLDw53etzNnzmjt2rVO731cXJw2bNjgKLNkyRLZ7XbHF2l0dLRWrFih5ORkR5mFCxeqYsWKKlSokKMM7ePG8s8//+jEiROKiIiQRFtwJ8YYDRo0SF9//bWWLFmS4XLM6/W9wO+O/OFK7SEzmzZtkiSnzwfagwu5urcLpJk9e7bx9fU1M2bMMNu3bzf9+/c3ISEhTr2w4Mbz5JNPmmXLlpl9+/aZVatWmSZNmpiiRYuao0ePGmPSurktWbKkWbJkiVm/fr2Jjo420dHRjuXTuzVt1qyZ2bRpk1mwYIEpVqxYpt2aDh8+3OzYscO8/fbbdH+eD5w9e9Zs3LjRbNy40UgyEydONBs3bjR///23MSat+/OQkBDzzTffmC1btpg2bdpk2v35bbfdZtauXWt++eUXU758eacur+Pi4kxYWJjp3r272bp1q5k9e7YpUKBAhi6vvby8zGuvvWZ27NhhRo4cSZfX11l2beHs2bNm2LBhZs2aNWbfvn1m0aJFplatWqZ8+fImISHBUQdtwT08+uijJjg42CxbtsypO+vz5887ylyv7wV+d7jeldrDnj17zIsvvmjWr19v9u3bZ7755htTpkwZ06BBA0cdtAfXIpHKR958801TsmRJ4+PjY+644w7z66+/ujokXKMHH3zQREREGB8fH3PLLbeYBx980OzZs8cx/8KFC2bAgAGmUKFCpkCBAqZdu3bm8OHDTnXs37/ftGjRwvj7+5uiRYuaJ5980iQnJzuVWbp0qbn11luNj4+PKVOmjJk+ffr12DxkY+nSpUZShqFHjx7GmLQu0J9//nkTFhZmfH19TePGjc2uXbuc6jhx4oTp0qWLCQgIMEFBQaZXr17m7NmzTmU2b95s6tWrZ3x9fc0tt9xixo0blyGWL774wlSoUMH4+PiYqlWrmu+//z7PthsZZdcWzp8/b5o1a2aKFStmvL29TVRUlOnXr1+GHy+0BfeQWTuQ5PSZfT2/F/jd4VpXag8HDhwwDRo0MIULFza+vr6mXLlyZvjw4U7PkTKG9uBKNmOMuX7nvwAAAADgxsc9UgAAAABgEYkUAAAAAFhEIgUAAAAAFpFIAQAAAIBFJFIAAAAAYBGJFAAAAABYRCIFAAAAABaRSAEAAACARSRSAHCDstlsmjdvXpbzS5UqpcmTJ+e4vhkzZigkJCTbMqNGjdKtt96a4zrzM6v750bx/PPPq3///q4OI1sLFizQrbfeKrvd7upQAOCqkUgBQD507NgxPfrooypZsqR8fX0VHh6umJgYrVq1Ksd1rFu3Ll/+oG7UqJEGDx7s6jDy7f65FrGxsZoyZYqeffZZV4eSrebNm8vb21uffPKJq0MBgKvm5eoAAAAZdejQQUlJSZo5c6bKlCmjI0eOaPHixTpx4kSO6yhWrFgeRuh6SUlJ8vHxcZpmjFFqaqq8vK789eaK/WMlvqvxwQcf6K677lJUVFSe1C9lvt+vRs+ePfXGG2+oe/fuuRAVAFx/nJECgHwmLi5OK1eu1Kuvvqp77rlHUVFRuuOOOzRixAjdf//9WS43cuRIRUREaMuWLZIyXro2ceJEVa9eXQULFlRkZKQGDBig+Pj4DPXMmzdP5cuXl5+fn2JiYnTw4MFs4/3ggw9UuXJl+fn5qVKlSpo6dWqWZXv27Knly5drypQpstlsstls2r9/vyRp69atatGihQICAhQWFqbu3bvr+PHjjmUbNWqkQYMGafDgwSpatKhiYmK0bNky2Ww2/fjjj6pdu7Z8fX31yy+/aO/evWrTpo3CwsIUEBCg22+/XYsWLXKK5fL9Y7PZ9MEHH6hdu3YqUKCAypcvr/nz52e77R999JHq1KmjwMBAhYeHq2vXrjp69Khjflbx2e12jR07VqVLl5a/v79q1qypL7/80rFcamqq+vTp45hfsWJFTZkyJdtYJGn27Nlq3bq107QFCxaoXr16CgkJUZEiRXTfffdp7969TmX++ecfdenSRYULF1bBggVVp04drV27VtLFyzk/+OADlS5dWn5+fpLS2mnfvn1VrFgxBQUF6d5779XmzZsddW7evFn33HOPAgMDFRQUpNq1a2v9+vWO+a1bt9b69eszxAIANwoSKQDIZwICAhQQEKB58+YpMTHxiuWNMXrsscc0a9YsrVy5UjVq1Mi0nIeHh9544w1t27ZNM2fO1JIlS/TUU085lTl//rxefvllzZo1S6tWrVJcXJw6d+6c5bo/+eQTvfDCC3r55Ze1Y8cOvfLKK3r++ec1c+bMTMtPmTJF0dHR6tevnw4fPqzDhw8rMjJScXFxuvfee3Xbbbdp/fr1WrBggY4cOaJOnTo5LT9z5kz5+Pho1apVmjZtmmP6M888o3HjxmnHjh2qUaOG4uPj1bJlSy1evFgbN25U8+bN1bp1ax04cCDbfTl69Gh16tRJW7ZsUcuWLdWtWzedPHkyy/LJyckaM2aMNm/erHnz5mn//v3q2bNnhnKXxzd27FjNmjVL06ZN07Zt2zRkyBA99NBDWr58uSTJbrerRIkSmjNnjrZv364XXnhB//3vf/XFF19kGcvJkye1fft21alTx2n6uXPnNHToUK1fv16LFy+Wh4eH2rVr57g/KT4+Xg0bNtS///6r+fPna/PmzXrqqaec7l/as2ePvvrqK82dO1ebNm2SJD3wwAM6evSofvzxR23YsEG1atVS48aNHfurW7duKlGihNatW6cNGzbomWeekbe3t6POkiVLKiwsTCtXrsz2PQGAfMsAAPKdL7/80hQqVMj4+fmZu+66y4wYMcJs3rzZqYwkM2fOHNO1a1dTuXJl888//zjNj4qKMpMmTcpyHXPmzDFFihRxvJ4+fbqRZH799VfHtB07dhhJZu3atcYYY0aOHGlq1qzpmF+2bFnz6aefOtU7ZswYEx0dneV6GzZsaJ544okMyzRr1sxp2sGDB40ks2vXLsdyt912m1OZpUuXGklm3rx5Wa4vXdWqVc2bb77peH35/pFknnvuOcfr+Ph4I8n8+OOPV6w73bp164wkc/bs2SzjS0hIMAUKFDCrV692WrZPnz6mS5cuWdY9cOBA06FDhyznb9y40UgyBw4cyDbGY8eOGUnmjz/+MMYY8+6775rAwEBz4sSJTMuPHDnSeHt7m6NHjzqmrVy50gQFBZmEhASnsmXLljXvvvuuMcaYwMBAM2PGjGxjue2228yoUaOyLQMA+RVnpAAgH+rQoYMOHTqk+fPnq3nz5lq2bJlq1aqlGTNmOJUbMmSI1q5dqxUrVuiWW27Jts5FixapcePGuuWWWxQYGKju3bvrxIkTOn/+vKOMl5eXbr/9dsfrSpUqKSQkRDt27MhQ37lz57R371716dPHcRYtICBAL730kuXLtTZv3qylS5c61VOpUiVJcqqrdu3amS5/+VmY+Ph4DRs2TJUrV1ZISIgCAgK0Y8eOK56RuvRsXsGCBRUUFOR0qd7lNmzYoNatW6tkyZIKDAxUw4YNJSnDei6Nb8+ePTp//ryaNm3qtL2zZs1y2ta3335btWvXVrFixRQQEKD33nsv2/gvXLggSY5L79Lt3r1bXbp0UZkyZRQUFKRSpUo5xbhp0ybddtttKly4cJZ1R0VFOd1TtnnzZsXHx6tIkSJO27Bv3z7HNgwdOlR9+/ZVkyZNNG7cuEzbhL+/v1P7A4AbCZ1NAEA+5efnp6ZNm6pp06Z6/vnn1bdvX40cOdLp0rGmTZvqs88+008//aRu3bplWdf+/ft133336dFHH9XLL7+swoUL65dfflGfPn2UlJSkAgUKWI4v/f6q999/X3Xr1nWa5+npabmu1q1b69VXX80wLyIiwjFesGDBTJe/fPqwYcO0cOFCvfbaaypXrpz8/f3VsWNHJSUlZRvHpZeeSWn3TWXVRfe5c+cUExOjmJgYffLJJypWrJgOHDigmJiYDOu5NL70/fb9999nSH59fX0lpd3rNGzYML3++uuKjo5WYGCgJkyY4LhvKTNFixaVJJ06dcop6WndurWioqL0/vvvq3jx4rLb7apWrZojRn9//2z3yeXxp29DRESEli1blqFsehf6o0aNUteuXfX999/rxx9/1MiRIzV79my1a9fOUfbkyZNu3ykKAPdFIgUAN4gqVapkeG7U/fffr9atW6tr167y9PTM8n6mDRs2yG636/XXX5eHR9rFCJndb5OSkqL169frjjvukCTt2rVLcXFxqly5coayYWFhKl68uP76669sk7jL+fj4KDU11WlarVq19NVXX6lUqVK50qPdqlWr1LNnT8eP9vj4eEenFrll586dOnHihMaNG6fIyEhJcupMIStVqlSRr6+vDhw44DiDdblVq1bprrvu0oABAxzTrnSWr2zZsgoKCtL27dtVoUIFSdKJEye0a9cuvf/++6pfv74k6ZdffnFarkaNGvrggw908uTJbM9KXapWrVqKjY2Vl5eX4wxXZipUqKAKFSpoyJAh6tKli6ZPn+54TxISErR3717ddtttOVonAOQ3XNoHAPnMiRMndO+99+rjjz/Wli1btG/fPs2ZM0fjx49XmzZtMpRv166dPvroI/Xq1cup57dLlStXTsnJyXrzzTf1119/6aOPPnLqrCGdt7e3HnvsMa1du1YbNmxQz549deeddzoSq8uNHj1aY8eO1RtvvKE///xTf/zxh6ZPn66JEydmuX2lSpXS2rVrtX//fh0/flx2u10DBw7UyZMn1aVLF61bt0579+7VTz/9pF69emVIunKifPnyjo4RNm/erK5du+b6w19LliwpHx8fxz6dP3++xowZc8XlAgMDNWzYMA0ZMkQzZ87U3r179fvvv+vNN990dNJRvnx5rV+/Xj/99JP+/PNPPf/881q3bl229Xp4eKhJkyZOiVKhQoVUpEgRvffee9qzZ4+WLFmioUOHOi3XpUsXhYeHq23btlq1apX++usvffXVV1qzZk2W62rSpImio6PVtm1b/fzzz9q/f79Wr16tZ599VuvXr9eFCxc0aNAgLVu2TH///bdWrVqldevWOSXkv/76q3x9fRUdHX3FfQYA+RGJFADkMwEBAapbt64mTZqkBg0aqFq1anr++efVr18/vfXWW5ku07FjR82cOVPdu3fX3LlzM8yvWbOmJk6cqFdffVXVqlXTJ598orFjx2YoV6BAAT399NPq2rWr7r77bgUEBOjzzz/PMta+ffvqgw8+0PTp01W9enU1bNhQM2bMUOnSpbNcZtiwYfL09FSVKlUcl8MVL15cq1atUmpqqpo1a6bq1atr8ODBCgkJcZxBs2LixIkqVKiQ7rrrLrVu3VoxMTGqVauW5XqyU6xYMc2YMUNz5sxRlSpVNG7cOL322ms5WnbMmDF6/vnnNXbsWFWuXFnNmzfX999/79hvjzzyiNq3b68HH3xQdevW1YkTJ5zOTmWlb9++mj17tiNp9PDw0OzZs7VhwwZVq1ZNQ4YM0YQJE5yW8fHx0c8//6zQ0FC1bNlS1atX17hx47K9PNNms+mHH35QgwYN1KtXL1WoUEGdO3fW33//rbCwMHl6eurEiRN6+OGHVaFCBXXq1EktWrTQ6NGjHXV89tln6tat21VdVgoA+YHNGGNcHQQAALh2xhjVrVvXcSldfnX8+HFVrFhR69evzzbpBoD8jDNSAAC4CZvNpvfee08pKSmuDiVb+/fv19SpU0miANzQOCMFAAAAABZxRgoAAAAALCKRAgAAAACLSKQAAAAAwCISKQAAAACwiEQKAAAAACwikQIAAAAAi0ikAAAAAMAiEikAAAAAsIhECgAAAAAs+j8i65ieTvVYLQAAAABJRU5ErkJggg==\n"},"metadata":{}}],"source":["plot_compare('SkiableTerrain_ac', 'Skiable terrain area (acres)')"]},{"cell_type":"markdown","metadata":{"id":"n7dVUVERXvBg"},"source":["Big Mountain is amongst the resorts with the largest amount of skiable terrain."]},{"cell_type":"markdown","metadata":{"id":"-zJEVNCxXvBg"},"source":["## 5.9 Modeling scenarios"]},{"cell_type":"markdown","metadata":{"id":"ed-viljDXvBg"},"source":["Big Mountain Resort has been reviewing potential scenarios for either cutting costs or increasing revenue (from ticket prices). Ticket price is not determined by any set of parameters; the resort is free to set whatever price it likes. However, the resort operates within a market where people pay more for certain facilities, and less for others. Being able to sense how facilities support a given ticket price is valuable business intelligence. This is where the utility of our model comes in.\n","\n","The business has shortlisted some options:\n","1. Permanently closing down up to 10 of the least used runs. This doesn't impact any other resort statistics.\n","2. Increase the vertical drop by adding a run to a point 150 feet lower down but requiring the installation of an additional chair lift to bring skiers back up, without additional snow making coverage\n","3. Same as number 2, but adding 2 acres of snow making cover\n","4. Increase the longest run by 0.2 mile to boast 3.5 miles length, requiring an additional snow making coverage of 4 acres\n","\n","The expected number of visitors over the season is 350,000 and, on average, visitors ski for five days. Assume the provided data includes the additional lift that Big Mountain recently installed."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"yOJrHvlFXvBg"},"outputs":[],"source":["expected_visitors = 350_000"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":81},"id":"60cXoz4SXvBg","executionInfo":{"status":"ok","timestamp":1721139258813,"user_tz":240,"elapsed":176,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"69f60af2-3a54-46ea-dc1e-7a5c20428d80"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" vertical_drop Snow Making_ac total_chairs fastQuads Runs \\\n","151 2353 600.0 14 3 105.0 \n","\n"," LongestRun_mi trams SkiableTerrain_ac \n","151 3.3 0 3000.0 "],"text/html":["\n","
\n","
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
vertical_dropSnow Making_actotal_chairsfastQuadsRunsLongestRun_mitramsSkiableTerrain_ac
1512353600.0143105.03.303000.0
\n","
\n","
\n","\n","
\n"," \n","\n"," \n","\n"," \n","
\n","\n","\n","
\n","
\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"big_mountain[all_feats]\",\n \"rows\": 1,\n \"fields\": [\n {\n \"column\": \"vertical_drop\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 2353,\n \"max\": 2353,\n \"num_unique_values\": 1,\n \"samples\": [\n 2353\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Snow Making_ac\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 600.0,\n \"max\": 600.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 600.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"total_chairs\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 14,\n \"max\": 14,\n \"num_unique_values\": 1,\n \"samples\": [\n 14\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"fastQuads\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 3,\n \"max\": 3,\n \"num_unique_values\": 1,\n \"samples\": [\n 3\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Runs\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 105.0,\n \"max\": 105.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 105.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"LongestRun_mi\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 3.3,\n \"max\": 3.3,\n \"num_unique_values\": 1,\n \"samples\": [\n 3.3\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"trams\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 0,\n \"max\": 0,\n \"num_unique_values\": 1,\n \"samples\": [\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"SkiableTerrain_ac\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 3000.0,\n \"max\": 3000.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 3000.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":28}],"source":["all_feats = ['vertical_drop', 'Snow Making_ac', 'total_chairs', 'fastQuads',\n"," 'Runs', 'LongestRun_mi', 'trams', 'SkiableTerrain_ac']\n","big_mountain[all_feats]"]},{"cell_type":"code","execution_count":null,"metadata":{"id":"xM898tUJXvBh"},"outputs":[],"source":["#Code task 2#\n","#In this function, copy the Big Mountain data into a new data frame\n","#(Note we use .copy()!)\n","#And then for each feature, and each of its deltas (changes from the original),\n","#create the modified scenario dataframe (bm2) and make a ticket price prediction\n","#for it. The difference between the scenario's prediction and the current\n","#prediction is then calculated and returned.\n","#Complete the code to increment each feature by the associated delta\n","def predict_increase(features, deltas):\n"," \"\"\"Increase in modelled ticket price by applying delta to feature.\n","\n"," Arguments:\n"," features - list, names of the features in the ski_data dataframe to change\n"," deltas - list, the amounts by which to increase the values of the features\n","\n"," Outputs:\n"," Amount of increase in the predicted ticket price\n"," \"\"\"\n","\n"," bm2 = X_bm.copy()\n"," for f, d in zip(features, deltas):\n"," # Check if the column exists in the DataFrame before accessing it\n"," if f in bm2.columns:\n"," bm2[f] += d\n"," else:\n"," print(f\"Warning: Column '{f}' not found in the DataFrame.\")\n"," return model.predict(bm2).item() - model.predict(X_bm).item()"]},{"cell_type":"markdown","metadata":{"id":"J39S2AGJXvBh"},"source":["### 5.9.1 Scenario 1"]},{"cell_type":"markdown","metadata":{"id":"pujZUdEfXvBh"},"source":["Close up to 10 of the least used runs. The number of runs is the only parameter varying."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"E53LAJ4dXvBh","executionInfo":{"status":"ok","timestamp":1721139279210,"user_tz":240,"elapsed":169,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"e591213a-365f-443f-8012-e78d61f5f60b"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["[-1, -2, -3, -4, -5, -6, -7, -8, -9, -10]"]},"metadata":{},"execution_count":30}],"source":["[i for i in range(-1, -11, -1)]"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"RmAEkll8XvBh","executionInfo":{"status":"ok","timestamp":1721139283517,"user_tz":240,"elapsed":187,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"74a24e5e-5d0f-4e68-866b-09e6eb50ab86"},"outputs":[{"output_type":"stream","name":"stdout","text":["Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'Runs' not found in the DataFrame.\n"]}],"source":["runs_delta = [i for i in range(-1, -11, -1)]\n","price_deltas = [predict_increase(['Runs'], [delta]) for delta in runs_delta]"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"X5rkyuC8XvBh","executionInfo":{"status":"ok","timestamp":1721139286426,"user_tz":240,"elapsed":174,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"3d37d9f4-c1a3-4b40-95ee-df9167e1daee"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]"]},"metadata":{},"execution_count":32}],"source":["price_deltas"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":487},"id":"-cPER_saXvBi","executionInfo":{"status":"ok","timestamp":1721139289677,"user_tz":240,"elapsed":409,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"42c6a2e0-77d8-4fbb-fcc0-65fbd71e83da"},"outputs":[{"output_type":"display_data","data":{"text/plain":["
"],"image/png":"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\n"},"metadata":{}}],"source":["#Code task 3#\n","#Create two plots, side by side, for the predicted ticket price change (delta) for each\n","#condition (number of runs closed) in the scenario and the associated predicted revenue\n","#change on the assumption that each of the expected visitors buys 5 tickets\n","#There are two things to do here:\n","#1 - use a list comprehension to create a list of the number of runs closed from `runs_delta`\n","#2 - use a list comprehension to create a list of predicted revenue changes from `price_deltas`\n","runs_closed = [-1 * delta for delta in runs_delta] #1 Use delta instead of Runs\n","fig, ax = plt.subplots(1, 2, figsize=(10, 5))\n","fig.subplots_adjust(wspace=0.5)\n","ax[0].plot(runs_closed, price_deltas, 'o-')\n","ax[0].set(xlabel='Runs closed', ylabel='Change ($)', title='Ticket price')\n","revenue_deltas = [5 * expected_visitors * price for price in price_deltas] #2\n","ax[1].plot(runs_closed, revenue_deltas, 'o-')\n","ax[1].set(xlabel='Runs closed', ylabel='Change ($)', title='Revenue');"]},{"cell_type":"markdown","metadata":{"id":"gSXB2Kz7XvBi"},"source":["The model says closing one run makes no difference. Closing 2 and 3 successively reduces support for ticket price and so revenue. If Big Mountain closes down 3 runs, it seems they may as well close down 4 or 5 as there's no further loss in ticket price. Increasing the closures down to 6 or more leads to a large drop."]},{"cell_type":"markdown","metadata":{"id":"peYj8ZQLXvBi"},"source":["### 5.9.2 Scenario 2"]},{"cell_type":"markdown","metadata":{"id":"P05t5BJ9XvBi"},"source":["In this scenario, Big Mountain is adding a run, increasing the vertical drop by 150 feet, and installing an additional chair lift."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"PqASbMB-XvBi","executionInfo":{"status":"ok","timestamp":1721139327457,"user_tz":240,"elapsed":151,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"5dc99106-384d-4731-8bd6-58827b6f21a2"},"outputs":[{"output_type":"stream","name":"stdout","text":["Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'total_chairs' not found in the DataFrame.\n"]}],"source":["#Code task 4#\n","#Call `predict_increase` with a list of the features 'Runs', 'vertical_drop', and 'total_chairs'\n","#and associated deltas of 1, 150, and 1\n","ticket2_increase = predict_increase(['Runs', 'vertical_drop', 'total_chairs'], [1, 150, 1])\n","revenue2_increase = 5 * expected_visitors * ticket2_increase"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"1Bec-XgAXvBi","executionInfo":{"status":"ok","timestamp":1721139333071,"user_tz":240,"elapsed":155,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"77ada47e-3fab-46ed-9cbc-7286735ce2be"},"outputs":[{"output_type":"stream","name":"stdout","text":["This scenario increases support for ticket price by $2.25\n","Over the season, this could be expected to amount to $3931729\n"]}],"source":["print(f'This scenario increases support for ticket price by ${ticket2_increase:.2f}')\n","print(f'Over the season, this could be expected to amount to ${revenue2_increase:.0f}')"]},{"cell_type":"markdown","metadata":{"id":"PgjuaZ6UXvBj"},"source":["### 5.9.3 Scenario 3"]},{"cell_type":"markdown","metadata":{"id":"msoJZPDuXvBj"},"source":["In this scenario, you are repeating the previous one but adding 2 acres of snow making."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"R7twETEIXvBj","executionInfo":{"status":"ok","timestamp":1721139339060,"user_tz":240,"elapsed":159,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"966c783a-8c45-4e20-8cc9-6ee80ecafc99"},"outputs":[{"output_type":"stream","name":"stdout","text":["Warning: Column 'Runs' not found in the DataFrame.\n","Warning: Column 'total_chairs' not found in the DataFrame.\n","Warning: Column 'Snow Making_ac' not found in the DataFrame.\n"]}],"source":["#Code task 5#\n","#Repeat scenario 2 conditions, but add an increase of 2 to `Snow Making_ac`\n","ticket3_increase = predict_increase(['Runs', 'vertical_drop', 'total_chairs', 'Snow Making_ac'], [1, 150, 1, 2])\n","revenue3_increase = 5 * expected_visitors * ticket3_increase"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FFuGI-1_XvBj","executionInfo":{"status":"ok","timestamp":1721139354874,"user_tz":240,"elapsed":166,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"30ae6280-5833-4bb3-a928-5163162c71b8"},"outputs":[{"output_type":"stream","name":"stdout","text":["This scenario increases support for ticket price by $2.25\n","Over the season, this could be expected to amount to $3931729\n"]}],"source":["print(f'This scenario increases support for ticket price by ${ticket3_increase:.2f}')\n","print(f'Over the season, this could be expected to amount to ${revenue3_increase:.0f}')"]},{"cell_type":"markdown","metadata":{"id":"DJ9bJ8mRXvBj"},"source":["Such a small increase in the snow making area makes no difference!"]},{"cell_type":"markdown","metadata":{"id":"Ci4vvocBXvBj"},"source":["### 5.9.4 Scenario 4"]},{"cell_type":"markdown","metadata":{"id":"8FYua3GJXvBj"},"source":["This scenario calls for increasing the longest run by .2 miles and guaranteeing its snow coverage by adding 4 acres of snow making capability."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"XoK9KurgXvBk","executionInfo":{"status":"ok","timestamp":1721139474439,"user_tz":240,"elapsed":173,"user":{"displayName":"Jesse Lindsey","userId":"09886266696052659215"}},"outputId":"fbcb4fac-74d6-43a7-f713-f14a8d7c43d2"},"outputs":[{"output_type":"stream","name":"stdout","text":["Warning: Column 'LongestRun_mi' not found in the DataFrame.\n","Warning: Column 'Snow Making_ac' not found in the DataFrame.\n"]},{"output_type":"execute_result","data":{"text/plain":["0.0"]},"metadata":{},"execution_count":38}],"source":["#Code task 6#\n","#Predict the increase from adding 0.2 miles to `LongestRun_mi` and 4 to `Snow Making_ac`\n","predict_increase(['LongestRun_mi', 'Snow Making_ac'], [0.2, 4])"]},{"cell_type":"markdown","metadata":{"id":"13ZQzUIFXvBk"},"source":["No difference whatsoever. Although the longest run feature was used in the linear model, the random forest model (the one we chose because of its better performance) only has longest run way down in the feature importance list."]},{"cell_type":"markdown","metadata":{"id":"AT_pOGyaXvBk"},"source":["## 5.10 Summary"]},{"cell_type":"markdown","metadata":{"id":"N9aEGkAUXvBk"},"source":["**Q: 1** Write a summary of the results of modeling these scenarios. Start by starting the current position; how much does Big Mountain currently charge? What does your modelling suggest for a ticket price that could be supported in the marketplace by Big Mountain's facilities? How would you approach suggesting such a change to the business leadership? Discuss the additional operating cost of the new chair lift per ticket (on the basis of each visitor on average buying 5 day tickets) in the context of raising prices to cover this. For future improvements, state which, if any, of the modeled scenarios you'd recommend for further consideration. Suggest how the business might test, and progress, with any run closures."]},{"cell_type":"markdown","metadata":{"id":"9kPWSwcbXvBk"},"source":["**A: 1** Big Mountain Resort currently charges 81 dollars average price per ticket. The price suggested per ticket, from modeling the data, indicates an average ticket price of 82 dollars and 53 cents with a mean absolute error of roughly 14 dollars and 31 cents. This model/estimate surely indicates room for a ticket price increase. The modeling also indicates that adding a new chair lift can increase support for ticket price increase by about 2 dollars and 25 cents, which could be expected to amount in about 3 million 931 thousand 729 dollars over the season. Modeling also indicates that an increase in snow making area makes no difference. It seems the model also indicates that closing one run makes no difference. Closing 2 or 3 runs successively reduces support for a ticket price increase and of course revenue. Closing 4 or 5 indicates no further loss/gain in ticket price. Any amount of closures after 6 indicates a large drop in support for ticket price increase. I would recommend modeled scenario # 2, which is an increase in the vertical drop by adding a run to a point 150 feet lower down but requiring the installation of an additional chair lift to bring skiers back up, without additional snow making coverage."]},{"cell_type":"markdown","metadata":{"id":"f1PtQXc1XvBk"},"source":["## 5.11 Further work"]},{"cell_type":"markdown","metadata":{"id":"ccmAjvTvXvBk"},"source":["**Q: 2** What next? Highlight any deficiencies in the data that hampered or limited this work. The only price data in our dataset were ticket prices. You were provided with information about the additional operating cost of the new chair lift, but what other cost information would be useful? Big Mountain was already fairly high on some of the league charts of facilities offered, but why was its modeled price so much higher than its current price? Would this mismatch come as a surprise to the business executives? How would you find out? Assuming the business leaders felt this model was useful, how would the business make use of it? Would you expect them to come to you every time they wanted to test a new combination of parameters in a scenario? We hope you would have better things to do, so how might this model be made available for business analysts to use and explore?"]},{"cell_type":"markdown","metadata":{"id":"ocyFtfSwXvBk"},"source":["**A: 2** The ‘Runs’ data not being found in the DataFrame hampered/limited the findings in this assignment. That information and cost of each specific Run would be useful information for data understanding. The modeled price estimating so high compared to the actual price could be because of the possibility that some of the competing resorts are overpriced and Big Mountain Resort could be underpricing. Based off of the data and comparison of what is offered at Big Mountain and what is offered at other resorts, as well as, the estimated revenue increase for Big Mountain, I think that the business executives would be surprised and pleased with this information. They could make use of this information by saving the file and altering the searches/information based on what findings are being requested."]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.7.9"},"toc":{"base_numbering":1,"nav_menu":{},"number_sections":true,"sideBar":true,"skip_h1_title":false,"title_cell":"Table of Contents","title_sidebar":"Contents","toc_cell":false,"toc_position":{},"toc_section_display":true,"toc_window_display":true},"varInspector":{"cols":{"lenName":16,"lenType":16,"lenVar":40},"kernels_config":{"python":{"delete_cmd_postfix":"","delete_cmd_prefix":"del ","library":"var_list.py","varRefreshCmd":"print(var_dic_list())"},"r":{"delete_cmd_postfix":") ","delete_cmd_prefix":"rm(","library":"var_list.r","varRefreshCmd":"cat(var_dic_list()) "}},"types_to_exclude":["module","function","builtin_function_or_method","instance","_Feature"],"window_display":false},"colab":{"provenance":[{"file_id":"1VZUWflc8NeSSWJlQIJH33SkG2SQ5JaLj","timestamp":1721146759548}]}},"nbformat":4,"nbformat_minor":0} \ No newline at end of file diff --git a/copy-of-04_preprocessing_and_training.ipynb b/copy-of-04_preprocessing_and_training.ipynb new file mode 100644 index 000000000..896ca7414 --- /dev/null +++ b/copy-of-04_preprocessing_and_training.ipynb @@ -0,0 +1,5568 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3CoGE_woGC5a" + }, + "source": [ + "# 4 Pre-Processing and Training Data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "pTx0aRgFGC5b" + }, + "source": [ + "## 4.1 Contents\n", + "* [4 Pre-Processing and Training Data](#4_Pre-Processing_and_Training_Data)\n", + " * [4.1 Contents](#4.1_Contents)\n", + " * [4.2 Introduction](#4.2_Introduction)\n", + " * [4.3 Imports](#4.3_Imports)\n", + " * [4.4 Load Data](#4.4_Load_Data)\n", + " * [4.5 Extract Big Mountain Data](#4.5_Extract_Big_Mountain_Data)\n", + " * [4.6 Train/Test Split](#4.6_Train/Test_Split)\n", + " * [4.7 Initial Not-Even-A-Model](#4.7_Initial_Not-Even-A-Model)\n", + " * [4.7.1 Metrics](#4.7.1_Metrics)\n", + " * [4.7.1.1 R-squared, or coefficient of determination](#4.7.1.1_R-squared,_or_coefficient_of_determination)\n", + " * [4.7.1.2 Mean Absolute Error](#4.7.1.2_Mean_Absolute_Error)\n", + " * [4.7.1.3 Mean Squared Error](#4.7.1.3_Mean_Squared_Error)\n", + " * [4.7.2 sklearn metrics](#4.7.2_sklearn_metrics)\n", + " * [4.7.2.0.1 R-squared](#4.7.2.0.1_R-squared)\n", + " * [4.7.2.0.2 Mean absolute error](#4.7.2.0.2_Mean_absolute_error)\n", + " * [4.7.2.0.3 Mean squared error](#4.7.2.0.3_Mean_squared_error)\n", + " * [4.7.3 Note On Calculating Metrics](#4.7.3_Note_On_Calculating_Metrics)\n", + " * [4.8 Initial Models](#4.8_Initial_Models)\n", + " * [4.8.1 Imputing missing feature (predictor) values](#4.8.1_Imputing_missing_feature_(predictor)_values)\n", + " * [4.8.1.1 Impute missing values with median](#4.8.1.1_Impute_missing_values_with_median)\n", + " * [4.8.1.1.1 Learn the values to impute from the train set](#4.8.1.1.1_Learn_the_values_to_impute_from_the_train_set)\n", + " * [4.8.1.1.2 Apply the imputation to both train and test splits](#4.8.1.1.2_Apply_the_imputation_to_both_train_and_test_splits)\n", + " * [4.8.1.1.3 Scale the data](#4.8.1.1.3_Scale_the_data)\n", + " * [4.8.1.1.4 Train the model on the train split](#4.8.1.1.4_Train_the_model_on_the_train_split)\n", + " * [4.8.1.1.5 Make predictions using the model on both train and test splits](#4.8.1.1.5_Make_predictions_using_the_model_on_both_train_and_test_splits)\n", + " * [4.8.1.1.6 Assess model performance](#4.8.1.1.6_Assess_model_performance)\n", + " * [4.8.1.2 Impute missing values with the mean](#4.8.1.2_Impute_missing_values_with_the_mean)\n", + " * [4.8.1.2.1 Learn the values to impute from the train set](#4.8.1.2.1_Learn_the_values_to_impute_from_the_train_set)\n", + " * [4.8.1.2.2 Apply the imputation to both train and test splits](#4.8.1.2.2_Apply_the_imputation_to_both_train_and_test_splits)\n", + " * [4.8.1.2.3 Scale the data](#4.8.1.2.3_Scale_the_data)\n", + " * [4.8.1.2.4 Train the model on the train split](#4.8.1.2.4_Train_the_model_on_the_train_split)\n", + " * [4.8.1.2.5 Make predictions using the model on both train and test splits](#4.8.1.2.5_Make_predictions_using_the_model_on_both_train_and_test_splits)\n", + " * [4.8.1.2.6 Assess model performance](#4.8.1.2.6_Assess_model_performance)\n", + " * [4.8.2 Pipelines](#4.8.2_Pipelines)\n", + " * [4.8.2.1 Define the pipeline](#4.8.2.1_Define_the_pipeline)\n", + " * [4.8.2.2 Fit the pipeline](#4.8.2.2_Fit_the_pipeline)\n", + " * [4.8.2.3 Make predictions on the train and test sets](#4.8.2.3_Make_predictions_on_the_train_and_test_sets)\n", + " * [4.8.2.4 Assess performance](#4.8.2.4_Assess_performance)\n", + " * [4.9 Refining The Linear Model](#4.9_Refining_The_Linear_Model)\n", + " * [4.9.1 Define the pipeline](#4.9.1_Define_the_pipeline)\n", + " * [4.9.2 Fit the pipeline](#4.9.2_Fit_the_pipeline)\n", + " * [4.9.3 Assess performance on the train and test set](#4.9.3_Assess_performance_on_the_train_and_test_set)\n", + " * [4.9.4 Define a new pipeline to select a different number of features](#4.9.4_Define_a_new_pipeline_to_select_a_different_number_of_features)\n", + " * [4.9.5 Fit the pipeline](#4.9.5_Fit_the_pipeline)\n", + " * [4.9.6 Assess performance on train and test data](#4.9.6_Assess_performance_on_train_and_test_data)\n", + " * [4.9.7 Assessing performance using cross-validation](#4.9.7_Assessing_performance_using_cross-validation)\n", + " * [4.9.8 Hyperparameter search using GridSearchCV](#4.9.8_Hyperparameter_search_using_GridSearchCV)\n", + " * [4.10 Random Forest Model](#4.10_Random_Forest_Model)\n", + " * [4.10.1 Define the pipeline](#4.10.1_Define_the_pipeline)\n", + " * [4.10.2 Fit and assess performance using cross-validation](#4.10.2_Fit_and_assess_performance_using_cross-validation)\n", + " * [4.10.3 Hyperparameter search using GridSearchCV](#4.10.3_Hyperparameter_search_using_GridSearchCV)\n", + " * [4.11 Final Model Selection](#4.11_Final_Model_Selection)\n", + " * [4.11.1 Linear regression model performance](#4.11.1_Linear_regression_model_performance)\n", + " * [4.11.2 Random forest regression model performance](#4.11.2_Random_forest_regression_model_performance)\n", + " * [4.11.3 Conclusion](#4.11.3_Conclusion)\n", + " * [4.12 Data quantity assessment](#4.12_Data_quantity_assessment)\n", + " * [4.13 Save best model object from pipeline](#4.13_Save_best_model_object_from_pipeline)\n", + " * [4.14 Summary](#4.14_Summary)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PK1m4ElWGC5d" + }, + "source": [ + "## 4.2 Introduction" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hu7BiJLzGC5d" + }, + "source": [ + "In preceding notebooks, performed preliminary assessments of data quality and refined the question to be answered. You found a small number of data values that gave clear choices about whether to replace values or drop a whole row. You determined that predicting the adult weekend ticket price was your primary aim. You threw away records with missing price data, but not before making the most of the other available data to look for any patterns between the states. You didn't see any and decided to treat all states equally; the state label didn't seem to be particularly useful.\n", + "\n", + "In this notebook you'll start to build machine learning models. Before even starting with learning a machine learning model, however, start by considering how useful the mean value is as a predictor. This is more than just a pedagogical device. You never want to go to stakeholders with a machine learning model only to have the CEO point out that it performs worse than just guessing the average! Your first model is a baseline performance comparitor for any subsequent model. You then build up the process of efficiently and robustly creating and assessing models against it. The development we lay out may be little slower than in the real world, but this step of the capstone is definitely more than just instructional. It is good practice to build up an understanding that the machine learning pipelines you build work as expected. You can validate steps with your own functions for checking expected equivalence between, say, pandas and sklearn implementations." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Kws2GzlEGC5d" + }, + "source": [ + "## 4.3 Imports" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3ErzW-67GC5e" + }, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import os\n", + "import pickle\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn import __version__ as sklearn_version\n", + "from sklearn.decomposition import PCA\n", + "from sklearn.preprocessing import scale\n", + "from sklearn.model_selection import train_test_split, cross_validate, GridSearchCV, learning_curve\n", + "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n", + "from sklearn.dummy import DummyRegressor\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.ensemble import RandomForestRegressor\n", + "from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error\n", + "from sklearn.pipeline import make_pipeline\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn.feature_selection import SelectKBest, f_regression\n", + "import datetime\n", + "\n", + "from sb_utils import save_file" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "v7uXe1oRGC5f" + }, + "source": [ + "## 4.4 Load Data" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true, + "id": "hP4qHm72GC5f", + "outputId": "28209102-a56b-4908-9cb6-afd695066a40", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 896 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " 0 1 2 \\\n", + "Name Alyeska Resort Eaglecrest Ski Area Hilltop Ski Area \n", + "Region Alaska Alaska Alaska \n", + "state Alaska Alaska Alaska \n", + "summit_elev 3939 2600 2090 \n", + "vertical_drop 2500 1540 294 \n", + "base_elev 250 1200 1796 \n", + "trams 1 0 0 \n", + "fastEight 0.0 0.0 0.0 \n", + "fastSixes 0 0 0 \n", + "fastQuads 2 0 0 \n", + "quad 2 0 0 \n", + "triple 0 0 1 \n", + "double 0 4 0 \n", + "surface 2 0 2 \n", + "total_chairs 7 4 3 \n", + "Runs 76.0 36.0 13.0 \n", + "TerrainParks 2.0 1.0 1.0 \n", + "LongestRun_mi 1.0 2.0 1.0 \n", + "SkiableTerrain_ac 1610.0 640.0 30.0 \n", + "Snow Making_ac 113.0 60.0 30.0 \n", + "daysOpenLastYear 150.0 45.0 150.0 \n", + "yearsOpen 60.0 44.0 36.0 \n", + "averageSnowfall 669.0 350.0 69.0 \n", + "AdultWeekday 65.0 47.0 30.0 \n", + "AdultWeekend 85.0 53.0 34.0 \n", + "projectedDaysOpen 150.0 90.0 152.0 \n", + "NightSkiing_ac 550.0 NaN 30.0 \n", + "\n", + " 3 4 \n", + "Name Arizona Snowbowl Sunrise Park Resort \n", + "Region Arizona Arizona \n", + "state Arizona Arizona \n", + "summit_elev 11500 11100 \n", + "vertical_drop 2300 1800 \n", + "base_elev 9200 9200 \n", + "trams 0 0 \n", + "fastEight 0.0 NaN \n", + "fastSixes 1 0 \n", + "fastQuads 0 1 \n", + "quad 2 2 \n", + "triple 2 3 \n", + "double 1 1 \n", + "surface 2 0 \n", + "total_chairs 8 7 \n", + "Runs 55.0 65.0 \n", + "TerrainParks 4.0 2.0 \n", + "LongestRun_mi 2.0 1.2 \n", + "SkiableTerrain_ac 777.0 800.0 \n", + "Snow Making_ac 104.0 80.0 \n", + "daysOpenLastYear 122.0 115.0 \n", + "yearsOpen 81.0 49.0 \n", + "averageSnowfall 260.0 250.0 \n", + "AdultWeekday 89.0 74.0 \n", + "AdultWeekend 89.0 78.0 \n", + "projectedDaysOpen 122.0 104.0 \n", + "NightSkiing_ac NaN 80.0 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
01234
NameAlyeska ResortEaglecrest Ski AreaHilltop Ski AreaArizona SnowbowlSunrise Park Resort
RegionAlaskaAlaskaAlaskaArizonaArizona
stateAlaskaAlaskaAlaskaArizonaArizona
summit_elev3939260020901150011100
vertical_drop2500154029423001800
base_elev2501200179692009200
trams10000
fastEight0.00.00.00.0NaN
fastSixes00010
fastQuads20001
quad20022
triple00123
double04011
surface20220
total_chairs74387
Runs76.036.013.055.065.0
TerrainParks2.01.01.04.02.0
LongestRun_mi1.02.01.02.01.2
SkiableTerrain_ac1610.0640.030.0777.0800.0
Snow Making_ac113.060.030.0104.080.0
daysOpenLastYear150.045.0150.0122.0115.0
yearsOpen60.044.036.081.049.0
averageSnowfall669.0350.069.0260.0250.0
AdultWeekday65.047.030.089.074.0
AdultWeekend85.053.034.089.078.0
projectedDaysOpen150.090.0152.0122.0104.0
NightSkiing_ac550.0NaN30.0NaN80.0
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "ski_data" + } + }, + "metadata": {}, + "execution_count": 3 + } + ], + "source": [ + "ski_data = pd.read_csv('https://raw.githubusercontent.com/springboard-curriculum/DataScienceGuidedCapstone/master/raw_data/ski_resort_data.csv')\n", + "ski_data.head().T" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DF3ErcM2GC5g" + }, + "source": [ + "## 4.5 Extract Big Mountain Data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "a1ztZZJaGC5g" + }, + "source": [ + "Big Mountain is your resort. Separate it from the rest of the data to use later." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Jx3b3TsqGC5g" + }, + "outputs": [], + "source": [ + "big_mountain = ski_data[ski_data.Name == 'Big Mountain Resort']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "C9EQfETrGC5g", + "outputId": "a35ecd1d-f6dd-4f24-fb1c-040731f9991f", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 896 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " 151\n", + "Name Big Mountain Resort\n", + "Region Montana\n", + "state Montana\n", + "summit_elev 6817\n", + "vertical_drop 2353\n", + "base_elev 4464\n", + "trams 0\n", + "fastEight 0.0\n", + "fastSixes 0\n", + "fastQuads 3\n", + "quad 2\n", + "triple 6\n", + "double 0\n", + "surface 3\n", + "total_chairs 14\n", + "Runs 105.0\n", + "TerrainParks 4.0\n", + "LongestRun_mi 3.3\n", + "SkiableTerrain_ac 3000.0\n", + "Snow Making_ac 600.0\n", + "daysOpenLastYear 123.0\n", + "yearsOpen 72.0\n", + "averageSnowfall 333.0\n", + "AdultWeekday 81.0\n", + "AdultWeekend 81.0\n", + "projectedDaysOpen 123.0\n", + "NightSkiing_ac 600.0" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
151
NameBig Mountain Resort
RegionMontana
stateMontana
summit_elev6817
vertical_drop2353
base_elev4464
trams0
fastEight0.0
fastSixes0
fastQuads3
quad2
triple6
double0
surface3
total_chairs14
Runs105.0
TerrainParks4.0
LongestRun_mi3.3
SkiableTerrain_ac3000.0
Snow Making_ac600.0
daysOpenLastYear123.0
yearsOpen72.0
averageSnowfall333.0
AdultWeekday81.0
AdultWeekend81.0
projectedDaysOpen123.0
NightSkiing_ac600.0
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"big_mountain\",\n \"rows\": 27,\n \"fields\": [\n {\n \"column\": 151,\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 19,\n \"samples\": [\n \"Big Mountain Resort\",\n 0,\n 4.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 5 + } + ], + "source": [ + "big_mountain.T" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "UdrDpfazGC5h", + "outputId": "edde232b-3d0c-4ce7-e930-79f9196de32c", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(330, 27)" + ] + }, + "metadata": {}, + "execution_count": 6 + } + ], + "source": [ + "ski_data.shape" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "BHdNVSFJGC5h" + }, + "outputs": [], + "source": [ + "ski_data = ski_data[ski_data.Name != 'Big Mountain Resort']" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qcegR2OiGC5h", + "outputId": "544eb919-2143-4917-ab57-df20ec44621a", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(329, 27)" + ] + }, + "metadata": {}, + "execution_count": 8 + } + ], + "source": [ + "ski_data.shape" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "w79lyK8DGC5i" + }, + "source": [ + "## 4.6 Train/Test Split" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "yquy-0UNGC5i" + }, + "source": [ + "So far, you've treated ski resort data as a single entity. In machine learning, when you train your model on all of your data, you end up with no data set aside to evaluate model performance. You could keep making more and more complex models that fit the data better and better and not realise you were overfitting to that one set of samples. By partitioning the data into training and testing splits, without letting a model (or missing-value imputation) learn anything about the test split, you have a somewhat independent assessment of how your model might perform in the future. An often overlooked subtlety here is that people all too frequently use the test set to assess model performance _and then compare multiple models to pick the best_. This means their overall model selection process is fitting to one specific data set, now the test split. You could keep going, trying to get better and better performance on that one data set, but that's where cross-validation becomes especially useful. While training models, a test split is very useful as a final check on expected future performance." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vd2AHm-5GC5i" + }, + "source": [ + "What partition sizes would you have with a 70/30 train/test split?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "P117wBQ7GC5i", + "outputId": "f1241f44-b67f-4718-c469-5186b2e4c11c", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(230.29999999999998, 98.7)" + ] + }, + "metadata": {}, + "execution_count": 9 + } + ], + "source": [ + "len(ski_data) * .7, len(ski_data) * .3" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Vz_1LHNDGC5j" + }, + "outputs": [], + "source": [ + "X_train, X_test, y_train, y_test = train_test_split(ski_data.drop(columns='AdultWeekend'),\n", + " ski_data.AdultWeekend, test_size=0.3,\n", + " random_state=47)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "BZLkeJtlGC5j", + "outputId": "383805da-9062-4f5a-8764-e75591a5bdf5", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "((230, 26), (99, 26))" + ] + }, + "metadata": {}, + "execution_count": 11 + } + ], + "source": [ + "X_train.shape, X_test.shape" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "OzYM68KFGC5j", + "outputId": "99f064d0-41a5-4dcc-e096-0eaf131de96c", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "((230,), (99,))" + ] + }, + "metadata": {}, + "execution_count": 12 + } + ], + "source": [ + "y_train.shape, y_test.shape" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "kpTR4dddGC5j" + }, + "outputs": [], + "source": [ + "#Code task 1#\n", + "#Save the 'Name', 'state', and 'Region' columns from the train/test data into names_train and names_test\n", + "#Then drop those columns from `X_train` and `X_test`. Use 'inplace=True'\n", + "names_list = ['Name', 'state', 'Region']\n", + "names_train = X_train[names_list]\n", + "names_test = X_test[names_list]\n", + "X_train.drop(columns=names_list, inplace=True)\n", + "X_test.drop(columns=names_list, inplace=True)\n", + "X_train.shape, X_test.shape" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "IxoNlzjcGC5k", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "7abc004e-214d-42b3-849b-4e723355f750" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Name object\n", + "Region object\n", + "state object\n", + "summit_elev int64\n", + "vertical_drop int64\n", + "base_elev int64\n", + "trams int64\n", + "fastEight float64\n", + "fastSixes int64\n", + "fastQuads int64\n", + "quad int64\n", + "triple int64\n", + "double int64\n", + "surface int64\n", + "total_chairs int64\n", + "Runs float64\n", + "TerrainParks float64\n", + "LongestRun_mi float64\n", + "SkiableTerrain_ac float64\n", + "Snow Making_ac float64\n", + "daysOpenLastYear float64\n", + "yearsOpen float64\n", + "averageSnowfall float64\n", + "AdultWeekday float64\n", + "projectedDaysOpen float64\n", + "NightSkiing_ac float64\n", + "dtype: object" + ] + }, + "metadata": {}, + "execution_count": 13 + } + ], + "source": [ + "#Code task 2#\n", + "#Check the `dtypes` attribute of `X_train` to verify all features are numeric\n", + "X_train.dtypes" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "7r4MaiNiGC5k", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "2c7d88d7-7408-466b-bb31-cbeff151d7bd" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Name object\n", + "Region object\n", + "state object\n", + "summit_elev int64\n", + "vertical_drop int64\n", + "base_elev int64\n", + "trams int64\n", + "fastEight float64\n", + "fastSixes int64\n", + "fastQuads int64\n", + "quad int64\n", + "triple int64\n", + "double int64\n", + "surface int64\n", + "total_chairs int64\n", + "Runs float64\n", + "TerrainParks float64\n", + "LongestRun_mi float64\n", + "SkiableTerrain_ac float64\n", + "Snow Making_ac float64\n", + "daysOpenLastYear float64\n", + "yearsOpen float64\n", + "averageSnowfall float64\n", + "AdultWeekday float64\n", + "projectedDaysOpen float64\n", + "NightSkiing_ac float64\n", + "dtype: object" + ] + }, + "metadata": {}, + "execution_count": 14 + } + ], + "source": [ + "#Code task 3#\n", + "#Repeat this check for the test split in `X_test`\n", + "X_test.dtypes" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Br24BVWSGC5k" + }, + "source": [ + "You have only numeric features in your X now!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "d8gYY4UqGC5k" + }, + "source": [ + "## 4.7 Initial Not-Even-A-Model" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iaIhqjRYGC5k" + }, + "source": [ + "A good place to start is to see how good the mean is as a predictor. In other words, what if you simply say your best guess is the average price?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "DjCfyFmOGC5k", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "068d847a-5277-4cc3-94ff-4f0b6160f94f" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "64.5370618556701" + ] + }, + "metadata": {}, + "execution_count": 15 + } + ], + "source": [ + "#Code task 4#\n", + "#Calculate the mean of `y_train`\n", + "train_mean = y_train.mean()\n", + "train_mean" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QZ6EBcHMGC5l" + }, + "source": [ + "`sklearn`'s `DummyRegressor` easily does this:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "GmGGBUE8GC5l", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "571b0cfa-b00a-41e3-d592-a30b1436c272" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([[64.53706186]])" + ] + }, + "metadata": {}, + "execution_count": 21 + } + ], + "source": [ + "#Code task 5#\n", + "#Fit the dummy regressor on the training data\n", + "#Hint, call its `.fit()` method with `X_train` and `y_train` as arguments\n", + "#Then print the object's `constant_` attribute and verify it's the same as the mean above\n", + "dumb_reg = DummyRegressor(strategy='mean')\n", + "y_train_clean = y_train.dropna() # Remove rows with NaN values in y_train\n", + "X_train_clean = X_train.loc[y_train_clean.index] # Subset X_train to match the non-NaN rows\n", + "dumb_reg.fit(X_train_clean, y_train_clean)\n", + "dumb_reg.constant_" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "i-oafl2cGC5l" + }, + "source": [ + "How good is this? How closely does this match, or explain, the actual values? There are many ways of assessing how good one set of values agrees with another, which brings us to the subject of metrics." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "K77RpsSsGC5l" + }, + "source": [ + "### 4.7.1 Metrics" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "o25DjTzvGC5l" + }, + "source": [ + "#### 4.7.1.1 R-squared, or coefficient of determination" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "B6iM0dMFGC5l" + }, + "source": [ + "One measure is $R^2$, the [coefficient of determination](https://en.wikipedia.org/wiki/Coefficient_of_determination). This is a measure of the proportion of variance in the dependent variable (our ticket price) that is predicted by our \"model\". The linked Wikipedia articles gives a nice explanation of how negative values can arise. This is frequently a cause of confusion for newcomers who, reasonably, ask how can a squared value be negative?\n", + "\n", + "Recall the mean can be denoted by $\\bar{y}$, where\n", + "\n", + "$$\\bar{y} = \\frac{1}{n}\\sum_{i=1}^ny_i$$\n", + "\n", + "and where $y_i$ are the individual values of the dependent variable.\n", + "\n", + "The total sum of squares (error), can be expressed as\n", + "\n", + "$$SS_{tot} = \\sum_i(y_i-\\bar{y})^2$$\n", + "\n", + "The above formula should be familiar as it's simply the variance without the denominator to scale (divide) by the sample size.\n", + "\n", + "The residual sum of squares is similarly defined to be\n", + "\n", + "$$SS_{res} = \\sum_i(y_i-\\hat{y})^2$$\n", + "\n", + "where $\\hat{y}$ are our predicted values for the depended variable.\n", + "\n", + "The coefficient of determination, $R^2$, here is given by\n", + "\n", + "$$R^2 = 1 - \\frac{SS_{res}}{SS_{tot}}$$\n", + "\n", + "Putting it into words, it's one minus the ratio of the residual variance to the original variance. Thus, the baseline model here, which always predicts $\\bar{y}$, should give $R^2=0$. A model that perfectly predicts the observed values would have no residual error and so give $R^2=1$. Models that do worse than predicting the mean will have increased the sum of squares of residuals and so produce a negative $R^2$." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "34Q6oEucGC5m" + }, + "outputs": [], + "source": [ + "#Code task 6#\n", + "#Calculate the R^2 as defined above\n", + "def r_squared(y, ypred):\n", + " \"\"\"R-squared score.\n", + "\n", + " Calculate the R-squared, or coefficient of determination, of the input.\n", + "\n", + " Arguments:\n", + " y -- the observed values\n", + " ypred -- the predicted values\n", + " \"\"\"\n", + " ybar = np.sum(y) / len(y) #yes, we could use np.mean(y)\n", + " sum_sq_tot = np.mean((y - ybar)**2) #total sum of squares error\n", + " sum_sq_res = np.mean((y - ypred)**2) #residual sum of squares error\n", + " R2 = 1.0 - sum_sq_tot / sum_sq_res\n", + " return R2" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "n3_ad9CbGC5m" + }, + "source": [ + "Make your predictions by creating an array of length the size of the training set with the single value of the mean." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "L9YDw26-GC5m", + "outputId": "f94a92eb-565a-4fa2-9b69-914c5fd95a39", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([64.53706186, 64.53706186, 64.53706186, 64.53706186, 64.53706186])" + ] + }, + "metadata": {}, + "execution_count": 23 + } + ], + "source": [ + "y_tr_pred_ = train_mean * np.ones(len(y_train))\n", + "y_tr_pred_[:5]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "WzwgNQGRGC5m" + }, + "source": [ + "Remember the `sklearn` dummy regressor?" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "oTjPeahxGC5s", + "outputId": "77dc6c54-b9a9-43bc-8b2f-54ca3fbf1b68", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([64.53706186, 64.53706186, 64.53706186, 64.53706186, 64.53706186])" + ] + }, + "metadata": {}, + "execution_count": 24 + } + ], + "source": [ + "y_tr_pred = dumb_reg.predict(X_train)\n", + "y_tr_pred[:5]" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BOXnR9XrGC5s" + }, + "source": [ + "You can see that `DummyRegressor` produces exactly the same results and saves you having to mess about broadcasting the mean (or whichever other statistic we used - check out the [documentation](https://scikit-learn.org/stable/modules/generated/sklearn.dummy.DummyRegressor.html) to see what's available) to an array of the appropriate length. It also gives you an object with `fit()` and `predict()` methods as well so you can use them as conveniently as any other `sklearn` estimator." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3Md1ngf1GC5s", + "outputId": "4d507e71-e552-4392-b419-ce4fa7ecebbc", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "-0.15644844476481623" + ] + }, + "metadata": {}, + "execution_count": 25 + } + ], + "source": [ + "r_squared(y_train, y_tr_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "RPcxUSB1GC5t" + }, + "source": [ + "Exactly as expected, if you use the average value as your prediction, you get an $R^2$ of zero _on our training set_. What if you use this \"model\" to predict unseen values from the test set? Remember, of course, that your \"model\" is trained on the training set; you still use the training set mean as your prediction." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "kWMVJH_DGC5t" + }, + "source": [ + "Make your predictions by creating an array of length the size of the test set with the single value of the (training) mean." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "xmmYDSFVGC5t", + "outputId": "d90052bc-fe12-4907-e539-8fbc292f93e1", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "-0.18384542904650614" + ] + }, + "metadata": {}, + "execution_count": 26 + } + ], + "source": [ + "y_te_pred = train_mean * np.ones(len(y_test))\n", + "r_squared(y_test, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "_iTc1viEGC5t" + }, + "source": [ + "Generally, you can expect performance on a test set to be slightly worse than on the training set. As you are getting an $R^2$ of zero on the training set, there's nowhere to go but negative!" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "w3p4qIgRGC5t" + }, + "source": [ + "$R^2$ is a common metric, and interpretable in terms of the amount of variance explained, it's less appealing if you want an idea of how \"close\" your predictions are to the true values. Metrics that summarise the difference between predicted and actual values are _mean absolute error_ and _mean squared error_." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "N3_-S015GC5u" + }, + "source": [ + "#### 4.7.1.2 Mean Absolute Error" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "42lbTNmjGC5u" + }, + "source": [ + "This is very simply the average of the absolute errors:\n", + "\n", + "$$MAE = \\frac{1}{n}\\sum_i^n|y_i - \\hat{y}|$$" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "0UYJMVAYGC5u" + }, + "outputs": [], + "source": [ + "#Code task 7#\n", + "#Calculate the MAE as defined above\n", + "def mae(y, ypred):\n", + " \"\"\"Mean absolute error.\n", + "\n", + " Calculate the mean absolute error of the arguments\n", + "\n", + " Arguments:\n", + " y -- the observed values\n", + " ypred -- the predicted values\n", + " \"\"\"\n", + " abs_error = np.abs(y_train - y_tr_pred)\n", + " mae = np.mean(abs_error)\n", + " return mae" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "bG4wIiGcGC5u", + "outputId": "45284b45-714e-4d15-d64d-7962d701aa46", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "18.717135189711982" + ] + }, + "metadata": {}, + "execution_count": 28 + } + ], + "source": [ + "mae(y_train, y_tr_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "rAj4GrUgGC5u", + "outputId": "8f846fae-f3b8-484b-a5c0-5d82123bcd54", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "18.717135189711982" + ] + }, + "metadata": {}, + "execution_count": 29 + } + ], + "source": [ + "mae(y_test, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "6WXCrqh5GC5v" + }, + "source": [ + "Mean absolute error is arguably the most intuitive of all the metrics, this essentially tells you that, on average, you might expect to be off by around \\\\$19 if you guessed ticket price based on an average of known values." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "-2a1BlzHGC5v" + }, + "source": [ + "#### 4.7.1.3 Mean Squared Error" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xOmD0rstGC5y" + }, + "source": [ + "Another common metric (and an important one internally for optimizing machine learning models) is the mean squared error. This is simply the average of the square of the errors:\n", + "\n", + "$$MSE = \\frac{1}{n}\\sum_i^n(y_i - \\hat{y})^2$$" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "scrolled": true, + "id": "fRvXO1RnGC5y" + }, + "outputs": [], + "source": [ + "#Code task 8#\n", + "#Calculate the MSE as defined above\n", + "def mse(y, ypred):\n", + " \"\"\"Mean square error.\n", + "\n", + " Calculate the mean square error of the arguments\n", + "\n", + " Arguments:\n", + " y -- the observed values\n", + " ypred -- the predicted values\n", + " \"\"\"\n", + " sq_error = (y_train - y_tr_pred)**2\n", + " mse = np.mean(sq_error)\n", + " return mse" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "szd0zqumGC5y", + "outputId": "1b337fd0-b2d3-417b-8ff5-ffa5614046f6", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "652.2235238415348" + ] + }, + "metadata": {}, + "execution_count": 31 + } + ], + "source": [ + "mse(y_train, y_tr_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "x5jEgX73GC5y", + "outputId": "77ffde96-527e-43bd-ad2c-746d4dfa4728", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "652.2235238415348" + ] + }, + "metadata": {}, + "execution_count": 32 + } + ], + "source": [ + "mse(y_test, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iNzlDw8oGC5z" + }, + "source": [ + "So here, you get a slightly better MSE on the test set than you did on the train set. And what does a squared error mean anyway? To convert this back to our measurement space, we often take the square root, to form the _root mean square error_ thus:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "4OkH5_MfGC5z", + "outputId": "f1cb4d24-3222-497e-977f-8325d4620d2e", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([25.53866723, 25.53866723])" + ] + }, + "metadata": {}, + "execution_count": 33 + } + ], + "source": [ + "np.sqrt([mse(y_train, y_tr_pred), mse(y_test, y_te_pred)])" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hJTztWA2GC5z" + }, + "source": [ + "### 4.7.2 sklearn metrics" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "kdqzOj0LGC5z" + }, + "source": [ + "Functions are good, but you don't want to have to define functions every time we want to assess performance. `sklearn.metrics` provides many commonly used metrics, included the ones above." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "tuxkdsw3GC5z" + }, + "source": [ + "##### 4.7.2.0.1 R-squared" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "yOKXPO87GC5z", + "outputId": "b7cf05bf-f703-4ce4-e7d5-6e2358a6d153", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "NaNs in y_train: True\n", + "NaNs in y_test: True\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.0, -0.0049471305923192155)" + ] + }, + "metadata": {}, + "execution_count": 37 + } + ], + "source": [ + "# Check for NaN values in y_train and y_test\n", + "print(\"NaNs in y_train:\", np.isnan(y_train).any())\n", + "print(\"NaNs in y_test:\", np.isnan(y_test).any())\n", + "\n", + "# If NaNs are present, handle them (e.g., impute or remove) before calling r2_score\n", + "# Example: Impute NaNs with the mean\n", + "y_train_imputed = np.nan_to_num(y_train, nan=np.nanmean(y_train))\n", + "y_test_imputed = np.nan_to_num(y_test, nan=np.nanmean(y_test))\n", + "\n", + "# Calculate R^2 scores using the imputed arrays\n", + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BSO4v_ILGC50" + }, + "source": [ + "##### 4.7.2.0.2 Mean absolute error" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3TmASnF0GC50", + "outputId": "340a6005-60dd-4046-ab22-db6ef2fd9524", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(15.787496638278801, 15.071073585635439)" + ] + }, + "metadata": {}, + "execution_count": 40 + } + ], + "source": [ + "mean_absolute_error(y_train_imputed, y_tr_pred), mean_absolute_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "nzd8tmhaGC50" + }, + "source": [ + "##### 4.7.2.0.3 Mean squared error" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "G4KfHJbVGC50", + "outputId": "503956c8-4983-437a-ce6a-0b1c1741d71e", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(550.1363635880772, 412.9315066406178)" + ] + }, + "metadata": {}, + "execution_count": 41 + } + ], + "source": [ + "mean_squared_error(y_train_imputed, y_tr_pred), mean_squared_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "DSHt-LEaGC50" + }, + "source": [ + "### 4.7.3 Note On Calculating Metrics" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "aPA8rBvJGC50" + }, + "source": [ + "When calling functions to calculate metrics, it is important to take care in the order of the arguments. Two of the metrics above actually don't care if the arguments are reversed; one does. Which one cares?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UBA1zcqdGC51" + }, + "source": [ + "In a Jupyter code cell, running `r2_score?` will bring up the docstring for the function, and `r2_score??` will bring up the actual code of the function! Try them and compare the source for `sklearn`'s function with yours. Feel free to explore what happens when you reverse the order of the arguments and compare behaviour of `sklearn`'s function and yours." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "cn2uK2SrGC51", + "outputId": "264287b3-795f-41c4-fc54-838606355222", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.0, 0.0)" + ] + }, + "metadata": {}, + "execution_count": 42 + } + ], + "source": [ + "# train set - sklearn\n", + "# correct order, incorrect order\n", + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_tr_pred, y_train_imputed)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "e8dq_UtmGC51", + "outputId": "1d362c42-485e-413d-c9fd-c863de943117", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(-0.0049471305923192155, 0.0)" + ] + }, + "metadata": {}, + "execution_count": 43 + } + ], + "source": [ + "# test set - sklearn\n", + "# correct order, incorrect order\n", + "r2_score(y_test_imputed, y_te_pred), r2_score(y_te_pred, y_test_imputed)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "hlS9kMhiGC51", + "outputId": "bd4d2a20-138a-42b6-990b-8f3ffd64aad7", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.0, 1.0)" + ] + }, + "metadata": {}, + "execution_count": 44 + } + ], + "source": [ + "# train set - using our homebrew function\n", + "# correct order, incorrect order\n", + "r_squared(y_train_imputed, y_tr_pred), r_squared(y_tr_pred, y_train_imputed)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "258N3g1PGC51", + "outputId": "b0ec29e1-ba4f-464f-9a7d-229a991b7205", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.004922776971763021, 1.0)" + ] + }, + "metadata": {}, + "execution_count": 45 + } + ], + "source": [ + "# test set - using our homebrew function\n", + "# correct order, incorrect order\n", + "r_squared(y_test_imputed, y_te_pred), r_squared(y_te_pred, y_test_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "2HtwxjroGC52" + }, + "source": [ + "You can get very different results swapping the argument order. It's worth highlighting this because data scientists do this too much in the real world! Don't be one of them! Frequently the argument order doesn't matter, but it will bite you when you do it with a function that does care. It's sloppy, bad practice and if you don't make a habit of putting arguments in the right order, you will forget!\n", + "\n", + "Remember:\n", + "* argument order matters,\n", + "* check function syntax with `func?` in a code cell" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "57JQhd7NGC52" + }, + "source": [ + "## 4.8 Initial Models" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "B1o4nWmUGC52" + }, + "source": [ + "### 4.8.1 Imputing missing feature (predictor) values" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GY-n5SFZGC52" + }, + "source": [ + "Recall when performing EDA, you imputed (filled in) some missing values in pandas. You did this judiciously for exploratory/visualization purposes. You left many missing values in the data. You can impute missing values using scikit-learn, but note that you should learn values to impute from a train split and apply that to the test split to then assess how well your imputation worked." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "O3T0zWoWGC52" + }, + "source": [ + "#### 4.8.1.1 Impute missing values with median" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MRXqJO8AGC52" + }, + "source": [ + "There's missing values. Recall from your data exploration that many distributions were skewed. Your first thought might be to impute missing values using the median." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8AUeJw1cGC52" + }, + "source": [ + "##### 4.8.1.1.1 Learn the values to impute from the train set" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "a4sqUn1qGC53", + "outputId": "42bf60f8-022f-402d-dfac-104c14a0583e", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "summit_elev 3075.0\n", + "vertical_drop 1000.0\n", + "base_elev 1491.5\n", + "trams 0.0\n", + "fastEight 0.0\n", + "fastSixes 0.0\n", + "fastQuads 0.0\n", + "quad 0.0\n", + "triple 1.0\n", + "double 1.0\n", + "surface 2.0\n", + "total_chairs 7.0\n", + "Runs 33.0\n", + "TerrainParks 2.0\n", + "LongestRun_mi 1.0\n", + "SkiableTerrain_ac 200.0\n", + "Snow Making_ac 102.5\n", + "daysOpenLastYear 110.0\n", + "yearsOpen 58.0\n", + "averageSnowfall 145.0\n", + "AdultWeekday 50.0\n", + "projectedDaysOpen 120.0\n", + "NightSkiing_ac 70.0\n", + "dtype: float64" + ] + }, + "metadata": {}, + "execution_count": 49 + } + ], + "source": [ + "# These are the values we'll use to fill in any missing values\n", + "# Handle non-numerical columns before calculating the median\n", + "X_train_numeric = X_train.select_dtypes(include=['number']) # Select only columns with numerical data\n", + "X_defaults_median = X_train_numeric.median()\n", + "X_defaults_median" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "2NjA4zUsGC53" + }, + "source": [ + "##### 4.8.1.1.2 Apply the imputation to both train and test splits" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "bNzmyKunGC53" + }, + "outputs": [], + "source": [ + "#Code task 9#\n", + "#Call `X_train` and `X_test`'s `fillna()` method, passing `X_defaults_median` as the values to use\n", + "#Assign the results to `X_tr` and `X_te`, respectively\n", + "X_tr = X_train.fillna(X_defaults_median)\n", + "X_te = X_test.fillna(X_defaults_median)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "pnTMgMmcGC53" + }, + "source": [ + "##### 4.8.1.1.3 Scale the data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PsZuEipQGC53" + }, + "source": [ + "As you have features measured in many different units, with numbers that vary by orders of magnitude, start off by scaling them to put them all on a consistent scale. The [StandardScaler](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html) scales each feature to zero mean and unit variance." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "vr8sSgU7GC53" + }, + "outputs": [], + "source": [ + "#Code task 10#\n", + "#Call the StandardScaler`s fit method on `X_tr` to fit the scaler\n", + "#then use it's `transform()` method to apply the scaling to both the train and test split\n", + "#data (`X_tr` and `X_te`), naming the results `X_tr_scaled` and `X_te_scaled`, respectively\n", + "# Drop the 'Region' column if it exists, as it is not a numerical feature. 'Name' has likely been dropped previously.\n", + "if 'Region' in X_tr.columns:\n", + " X_tr = X_tr.drop(['Region'], axis=1)\n", + "if 'Region' in X_te.columns:\n", + " X_te = X_te.drop(['Region'], axis=1)\n", + "\n", + "# Drop the 'state' column as it contains non-numerical values\n", + "if 'state' in X_tr.columns:\n", + " X_tr = X_tr.drop(['state'], axis=1)\n", + "if 'state' in X_te.columns:\n", + " X_te = X_te.drop(['state'], axis=1)\n", + "\n", + "scaler = StandardScaler()\n", + "scaler.fit(X_tr)\n", + "X_tr_scaled = scaler.transform(X_tr)\n", + "X_te_scaled = scaler.transform(X_te)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mez4eqv8GC54" + }, + "source": [ + "##### 4.8.1.1.4 Train the model on the train split" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "OHtcgRuEGC54" + }, + "outputs": [], + "source": [ + "lm = LinearRegression().fit(X_tr_scaled, y_train_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Ry_-iW7-GC54" + }, + "source": [ + "##### 4.8.1.1.5 Make predictions using the model on both train and test splits" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ohyybousGC54" + }, + "outputs": [], + "source": [ + "#Code task 11#\n", + "#Call the `predict()` method of the model (`lm`) on both the (scaled) train and test data\n", + "#Assign the predictions to `y_tr_pred` and `y_te_pred`, respectively\n", + "y_tr_pred = lm.predict(X_tr_scaled)\n", + "y_te_pred = lm.predict(X_te_scaled)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "3nUrhDpPGC54" + }, + "source": [ + "##### 4.8.1.1.6 Assess model performance" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mAiKt_jgGC54", + "outputId": "ee12a613-0664-4564-aefa-d68e06bda398", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.9041522057232028, 0.7976456921768976)" + ] + }, + "metadata": {}, + "execution_count": 61 + } + ], + "source": [ + "# r^2 - train, test\n", + "median_r2 = r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)\n", + "median_r2" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hPEux0TIGC55" + }, + "source": [ + "Recall that you estimated ticket price by simply using a known average. As expected, this produced an $R^2$ of zero for both the training and test set, because $R^2$ tells us how much of the variance you're explaining beyond that of using just the mean, and you were using just the mean. Here we see that our simple linear regression model explains over 80% of the variance on the train set and over 70% on the test set. Clearly you are onto something, although the much lower value for the test set suggests you're overfitting somewhat. This isn't a surprise as you've made no effort to select a parsimonious set of features or deal with multicollinearity in our data." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "S8w76dzZGC55", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "67b799c5-fcf5-48de-c706-9c66257dfe22" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(5.926090522538955, 6.3024374323465935)" + ] + }, + "metadata": {}, + "execution_count": 64 + } + ], + "source": [ + "#Code task 12#\n", + "#Now calculate the mean absolute error scores using `sklearn`'s `mean_absolute_error` function\n", + "# as we did above for R^2\n", + "# MAE - train, test\n", + "mae_score = mean_absolute_error\n", + "median_mae = mae_score(y_train_imputed, y_tr_pred), mae_score(y_test_imputed, y_te_pred)\n", + "median_mae" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "rGvbl4uIGC55" + }, + "source": [ + "Using this model, then, on average you'd expect to estimate a ticket price within \\\\$9 or so of the real price. This is much, much better than the \\\\$19 from just guessing using the average. There may be something to this machine learning lark after all!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "BxzkETKcGC55", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "a6d1bceb-1592-4640-dab8-e4922a72416d" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(52.72935700137537, 83.14712949661677)" + ] + }, + "metadata": {}, + "execution_count": 66 + } + ], + "source": [ + "#Code task 13#\n", + "#And also do the same using `sklearn`'s `mean_squared_error`\n", + "# MSE - train, test\n", + "mse_score = mean_squared_error\n", + "median_mse = mse_score(y_train_imputed, y_tr_pred), mse_score(y_test_imputed, y_te_pred)\n", + "median_mse" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jSszozHiGC55" + }, + "source": [ + "#### 4.8.1.2 Impute missing values with the mean" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Y56LkOw4GC55" + }, + "source": [ + "You chose to use the median for filling missing values because of the skew of many of our predictor feature distributions. What if you wanted to try something else, such as the mean?" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "KgKF4R3hGC56" + }, + "source": [ + "##### 4.8.1.2.1 Learn the values to impute from the train set" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "O2oS_cDXGC56", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "ed0262c8-7e23-4d8e-e455-7f0255ea8971" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "summit_elev 4592.652174\n", + "vertical_drop 1202.478261\n", + "base_elev 3390.656522\n", + "trams 0.169565\n", + "fastEight 0.008772\n", + "fastSixes 0.178261\n", + "fastQuads 1.008696\n", + "quad 0.913043\n", + "triple 1.500000\n", + "double 1.895652\n", + "surface 2.678261\n", + "total_chairs 8.347826\n", + "Runs 49.244541\n", + "TerrainParks 2.927835\n", + "LongestRun_mi 1.491150\n", + "SkiableTerrain_ac 642.676856\n", + "Snow Making_ac 188.234694\n", + "daysOpenLastYear 113.866667\n", + "yearsOpen 57.528384\n", + "averageSnowfall 177.734234\n", + "AdultWeekday 57.231526\n", + "projectedDaysOpen 120.944444\n", + "NightSkiing_ac 93.221374\n", + "dtype: float64" + ] + }, + "metadata": {}, + "execution_count": 69 + } + ], + "source": [ + "#Code task 14#\n", + "#As we did for the median above, calculate mean values for imputing missing values\n", + "# These are the values we'll use to fill in any missing values\n", + "X_defaults_mean = X_train.select_dtypes(include='number').mean() # Select only numeric columns\n", + "X_defaults_mean" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "aHLjx3vNGC56" + }, + "source": [ + "By eye, you can immediately tell that your replacement values are much higher than those from using the median." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "69UcNSjGGC56" + }, + "source": [ + "##### 4.8.1.2.2 Apply the imputation to both train and test splits" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "P_EGtIXFGC56" + }, + "outputs": [], + "source": [ + "X_tr = X_train.fillna(X_defaults_mean)\n", + "X_te = X_test.fillna(X_defaults_mean)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "kX_wvi3CGC56" + }, + "source": [ + "##### 4.8.1.2.3 Scale the data" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "VFlUjIBYGC57", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "7a9a608c-bc30-40f6-af0b-0c188e6804cd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "String columns in X_tr: Index(['Name', 'Region', 'state'], dtype='object')\n", + "String columns in X_te: Index(['Name', 'Region', 'state'], dtype='object')\n" + ] + } + ], + "source": [ + "# Identify columns with string values\n", + "string_columns_tr = X_tr.select_dtypes(include='object').columns\n", + "string_columns_te = X_te.select_dtypes(include='object').columns\n", + "\n", + "print(\"String columns in X_tr:\", string_columns_tr)\n", + "print(\"String columns in X_te:\", string_columns_te)\n", + "\n", + "# Decide how to handle these string columns:\n", + "# 1. Drop them if they are not relevant for scaling.\n", + "# 2. Encode them numerically using techniques like one-hot encoding or label encoding if they are relevant.\n", + "\n", + "# Example: Dropping string columns\n", + "X_tr_numeric = X_tr.drop(columns=string_columns_tr)\n", + "X_te_numeric = X_te.drop(columns=string_columns_te)\n", + "\n", + "# Now, try scaling again with the numeric data:\n", + "scaler = StandardScaler()\n", + "scaler.fit(X_tr_numeric)\n", + "X_tr_scaled = scaler.transform(X_tr_numeric)\n", + "X_te_scaled = scaler.transform(X_te_numeric)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "8tr7aO_QGC57" + }, + "source": [ + "##### 4.8.1.2.4 Train the model on the train split" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "tuFgNbUFGC57" + }, + "outputs": [], + "source": [ + "lm = LinearRegression().fit(X_tr_scaled, y_train_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "i2IZ9qfpGC57" + }, + "source": [ + "##### 4.8.1.2.5 Make predictions using the model on both train and test splits" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qpCTuFArGC57" + }, + "outputs": [], + "source": [ + "y_tr_pred = lm.predict(X_tr_scaled)\n", + "y_te_pred = lm.predict(X_te_scaled)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "C6Kp-72xGC57" + }, + "source": [ + "##### 4.8.1.2.6 Assess model performance" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gYlCy35UGC58", + "outputId": "0afb35e8-af23-4379-ff2a-f82f64ab8354", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.90228298680991, 0.8300429694208216)" + ] + }, + "metadata": {}, + "execution_count": 76 + } + ], + "source": [ + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "8Qw2218pGC58", + "outputId": "62b76561-332b-4972-e0d0-0f10ec64f171", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(5.758652911088192, 5.990687179884243)" + ] + }, + "metadata": {}, + "execution_count": 77 + } + ], + "source": [ + "mean_absolute_error(y_train_imputed, y_tr_pred), mean_absolute_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "c3j287ojGC58", + "outputId": "fc0a455a-3d2e-4ccb-f7ba-1793ae181db1", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(53.75768229708425, 69.8351291971559)" + ] + }, + "metadata": {}, + "execution_count": 78 + } + ], + "source": [ + "mean_squared_error(y_train_imputed, y_tr_pred), mean_squared_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "EsNxnJt6GC59" + }, + "source": [ + "These results don't seem very different to when you used the median for imputing missing values. Perhaps it doesn't make much difference here. Maybe your overtraining dominates. Maybe other feature transformations, such as taking the log, would help. You could try with just a subset of features rather than using all of them as inputs.\n", + "\n", + "To perform the median/mean comparison, you copied and pasted a lot of code just to change the function for imputing missing values. It would make more sense to write a function that performed the sequence of steps:\n", + "1. impute missing values\n", + "2. scale the features\n", + "3. train a model\n", + "4. calculate model performance\n", + "\n", + "But these are common steps and `sklearn` provides something much better than writing custom functions." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xgBxLYLzGC59" + }, + "source": [ + "### 4.8.2 Pipelines" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "waQ90iXvGC59" + }, + "source": [ + "One of the most important and useful components of `sklearn` is the [pipeline](https://scikit-learn.org/stable/modules/generated/sklearn.pipeline.Pipeline.html). In place of `panda`'s `fillna` DataFrame method, there is `sklearn`'s `SimpleImputer`. Remember the first linear model above performed the steps:\n", + "\n", + "1. replace missing values with the median for each feature\n", + "2. scale the data to zero mean and unit variance\n", + "3. train a linear regression model\n", + "\n", + "and all these steps were trained on the train split and then applied to the test split for assessment.\n", + "\n", + "The pipeline below defines exactly those same steps. Crucially, the resultant `Pipeline` object has a `fit()` method and a `predict()` method, just like the `LinearRegression()` object itself. Just as you might create a linear regression model and train it with `.fit()` and predict with `.predict()`, you can wrap the entire process of imputing and feature scaling and regression in a single object you can train with `.fit()` and predict with `.predict()`. And that's basically a pipeline: a model on steroids." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IQkam7dWGC59" + }, + "source": [ + "#### 4.8.2.1 Define the pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "MnDG5M6yGC59" + }, + "outputs": [], + "source": [ + "pipe = make_pipeline(\n", + " SimpleImputer(strategy='median'),\n", + " StandardScaler(),\n", + " LinearRegression()\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3QeiOGYmGC59", + "outputId": "229268dc-e7a9-4a23-952d-825d375dd6f2", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 187 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "sklearn.pipeline.Pipeline" + ], + "text/html": [ + "
\n", + " 
sklearn.pipeline.Pipeline
def __init__(steps, *, memory=None, verbose=False)
/usr/local/lib/python3.10/dist-packages/sklearn/pipeline.pyPipeline of transforms with a final estimator.\n",
+              "\n",
+              "Sequentially apply a list of transforms and a final estimator.\n",
+              "Intermediate steps of the pipeline must be 'transforms', that is, they\n",
+              "must implement `fit` and `transform` methods.\n",
+              "The final estimator only needs to implement `fit`.\n",
+              "The transformers in the pipeline can be cached using ``memory`` argument.\n",
+              "\n",
+              "The purpose of the pipeline is to assemble several steps that can be\n",
+              "cross-validated together while setting different parameters. For this, it\n",
+              "enables setting parameters of the various steps using their names and the\n",
+              "parameter name separated by a `'__'`, as in the example below. A step's\n",
+              "estimator may be replaced entirely by setting the parameter with its name\n",
+              "to another estimator, or a transformer removed by setting it to\n",
+              "`'passthrough'` or `None`.\n",
+              "\n",
+              "Read more in the :ref:`User Guide <pipeline>`.\n",
+              "\n",
+              ".. versionadded:: 0.5\n",
+              "\n",
+              "Parameters\n",
+              "----------\n",
+              "steps : list of tuple\n",
+              "    List of (name, transform) tuples (implementing `fit`/`transform`) that\n",
+              "    are chained in sequential order. The last transform must be an\n",
+              "    estimator.\n",
+              "\n",
+              "memory : str or object with the joblib.Memory interface, default=None\n",
+              "    Used to cache the fitted transformers of the pipeline. By default,\n",
+              "    no caching is performed. If a string is given, it is the path to\n",
+              "    the caching directory. Enabling caching triggers a clone of\n",
+              "    the transformers before fitting. Therefore, the transformer\n",
+              "    instance given to the pipeline cannot be inspected\n",
+              "    directly. Use the attribute ``named_steps`` or ``steps`` to\n",
+              "    inspect estimators within the pipeline. Caching the\n",
+              "    transformers is advantageous when fitting is time consuming.\n",
+              "\n",
+              "verbose : bool, default=False\n",
+              "    If True, the time elapsed while fitting each step will be printed as it\n",
+              "    is completed.\n",
+              "\n",
+              "Attributes\n",
+              "----------\n",
+              "named_steps : :class:`~sklearn.utils.Bunch`\n",
+              "    Dictionary-like object, with the following attributes.\n",
+              "    Read-only attribute to access any step parameter by user given name.\n",
+              "    Keys are step names and values are steps parameters.\n",
+              "\n",
+              "classes_ : ndarray of shape (n_classes,)\n",
+              "    The classes labels. Only exist if the last step of the pipeline is a\n",
+              "    classifier.\n",
+              "\n",
+              "n_features_in_ : int\n",
+              "    Number of features seen during :term:`fit`. Only defined if the\n",
+              "    underlying first estimator in `steps` exposes such an attribute\n",
+              "    when fit.\n",
+              "\n",
+              "    .. versionadded:: 0.24\n",
+              "\n",
+              "feature_names_in_ : ndarray of shape (`n_features_in_`,)\n",
+              "    Names of features seen during :term:`fit`. Only defined if the\n",
+              "    underlying estimator exposes such an attribute when fit.\n",
+              "\n",
+              "    .. versionadded:: 1.0\n",
+              "\n",
+              "See Also\n",
+              "--------\n",
+              "make_pipeline : Convenience function for simplified pipeline construction.\n",
+              "\n",
+              "Examples\n",
+              "--------\n",
+              ">>> from sklearn.svm import SVC\n",
+              ">>> from sklearn.preprocessing import StandardScaler\n",
+              ">>> from sklearn.datasets import make_classification\n",
+              ">>> from sklearn.model_selection import train_test_split\n",
+              ">>> from sklearn.pipeline import Pipeline\n",
+              ">>> X, y = make_classification(random_state=0)\n",
+              ">>> X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
+              "...                                                     random_state=0)\n",
+              ">>> pipe = Pipeline([('scaler', StandardScaler()), ('svc', SVC())])\n",
+              ">>> # The pipeline can be used as any other estimator\n",
+              ">>> # and avoids leaking the test set into the train set\n",
+              ">>> pipe.fit(X_train, y_train)\n",
+              "Pipeline(steps=[('scaler', StandardScaler()), ('svc', SVC())])\n",
+              ">>> pipe.score(X_test, y_test)\n",
+              "0.88
\n", + " \n", + "
" + ] + }, + "metadata": {}, + "execution_count": 113 + } + ], + "source": [ + "type(pipe)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "sItLPdQvGC59", + "outputId": "03ecb1c2-febf-493d-e222-70141b78265e", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(True, True)" + ] + }, + "metadata": {}, + "execution_count": 114 + } + ], + "source": [ + "hasattr(pipe, 'fit'), hasattr(pipe, 'predict')" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "1eHYXg9xGC5-" + }, + "source": [ + "#### 4.8.2.2 Fit the pipeline" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "N5I0TCA8GC5-" + }, + "source": [ + "Here, a single call to the pipeline's `fit()` method combines the steps of learning the imputation (determining what values to use to fill the missing ones), the scaling (determining the mean to subtract and the variance to divide by), and then training the model. It does this all in the one call with the training data as arguments." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "mB3tTsf6GC5-", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 158 + }, + "outputId": "25865cff-6b0f-4c5d-fd83-534511494ab0" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Pipeline(steps=[('preprocessor',\n", + " ColumnTransformer(transformers=[('cat', OneHotEncoder(),\n", + " ['Name'])]))])" + ], + "text/html": [ + "
Pipeline(steps=[('preprocessor',\n",
+              "                 ColumnTransformer(transformers=[('cat', OneHotEncoder(),\n",
+              "                                                  ['Name'])]))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 121 + } + ], + "source": [ + "#Code task 15#\n", + "#Call the pipe's `fit()` method with `X_train` and `y_train` as arguments\n", + "# Assuming 'name' is the column with non-numeric values\n", + "\n", + "# Verify if 'Name' is a column in X_train (Note capitalization)\n", + "if 'Name' not in X_train.columns:\n", + " # If 'Name' is not found, identify the correct column name\n", + " print(\"Column 'Name' not found. Available columns are:\", X_train.columns)\n", + " # Replace 'Name' with the correct column name below\n", + " categorical_features = ['replace_with_correct_column_name']\n", + "else:\n", + " categorical_features = ['Name'] # Use 'Name' if it exists\n", + "\n", + "numerical_features = X_train.columns.difference(categorical_features)\n", + "\n", + "# ... (rest of your code remains unchanged)\n", + "\n", + "# Change the 'name' reference in the ColumnTransformer to 'Name'\n", + "from sklearn.compose import ColumnTransformer\n", + "from sklearn.preprocessing import OneHotEncoder\n", + "\n", + "preprocessor = ColumnTransformer(\n", + " transformers=[\n", + " ('cat', OneHotEncoder(), categorical_features), # Use categorical_features here\n", + " # ... other transformers ...\n", + " ])\n", + "\n", + "pipe = Pipeline(steps=[\n", + " ('preprocessor', preprocessor),\n", + " # ... other steps ...\n", + "])\n", + "\n", + "pipe.fit(X_train, y_train_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "69GWYGPVGC5-" + }, + "source": [ + "#### 4.8.2.3 Make predictions on the train and test sets" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "lTi5LLlsGC5-" + }, + "outputs": [], + "source": [ + "# Add an estimator as the last step in the pipeline\n", + "from sklearn.linear_model import LinearRegression # Or any other suitable estimator\n", + "from sklearn.preprocessing import OneHotEncoder\n", + "\n", + "# ... (rest of your code remains unchanged)\n", + "\n", + "# Handle unknown categories in the OneHotEncoder\n", + "preprocessor = ColumnTransformer(\n", + " transformers=[\n", + " ('cat', OneHotEncoder(handle_unknown='ignore'), categorical_features), # Handle unknown categories\n", + " # ... other transformers ...\n", + " ])\n", + "\n", + "pipe = Pipeline(steps=[\n", + " ('preprocessor', preprocessor),\n", + " ('estimator', LinearRegression()) # Add an estimator here\n", + "])\n", + "\n", + "pipe.fit(X_train, y_train_imputed)\n", + "\n", + "# Now you can make predictions\n", + "y_tr_pred = pipe.predict(X_train)\n", + "y_te_pred = pipe.predict(X_test) # This should now work without raising an error" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QZykrRkMGC5-" + }, + "source": [ + "#### 4.8.2.4 Assess performance" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "TvxwVwloGC5_", + "outputId": "3b04ed61-b6de-46f8-ad86-2eac24719c0b", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(1.0, -0.005864211086616056)" + ] + }, + "metadata": {}, + "execution_count": 125 + } + ], + "source": [ + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qT9h-qLXGC5_" + }, + "source": [ + "And compare with your earlier (non-pipeline) result:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "08zPBAeyGC5_", + "outputId": "23ab1c24-df6c-486c-b0db-b35f7b37a2b1", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.9041522057232028, 0.7976456921768976)" + ] + }, + "metadata": {}, + "execution_count": 126 + } + ], + "source": [ + "median_r2" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "eq9x2JbTGC5_", + "outputId": "6edc077f-0a40-42fc-f77f-5ca2c1829cce", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(4.38682906425801e-15, 15.076498452864431)" + ] + }, + "metadata": {}, + "execution_count": 127 + } + ], + "source": [ + "mean_absolute_error(y_train_imputed, y_tr_pred), mean_absolute_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "PIlfrCaOGC5_" + }, + "outputs": [], + "source": [ + "#Compare with your earlier result:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "RZsDnEJkGC5_", + "outputId": "09de590d-1dcc-4401-dc9e-46e28a59b269", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(5.926090522538955, 6.3024374323465935)" + ] + }, + "metadata": {}, + "execution_count": 130 + } + ], + "source": [ + "median_mae" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gljUh-4dGC6A", + "outputId": "def6ded4-342c-44fe-fa86-aa6dfc5b367a", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(8.209641142334843e-29, 413.3083338573864)" + ] + }, + "metadata": {}, + "execution_count": 132 + } + ], + "source": [ + "mean_squared_error(y_train_imputed, y_tr_pred), mean_squared_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "WWiA1XlEGC6A" + }, + "source": [ + "Compare with your earlier result:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "A3KXSq96GC6A", + "outputId": "e1db5807-3883-41e8-a96d-85b73d486ce2", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(52.72935700137537, 83.14712949661677)" + ] + }, + "metadata": {}, + "execution_count": 133 + } + ], + "source": [ + "median_mse" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "bq9mGKVvGC6A" + }, + "source": [ + "These results confirm the pipeline is doing exactly what's expected, and results are identical to your earlier steps. This allows you to move faster but with confidence." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4b35Kt4XGC6A" + }, + "source": [ + "## 4.9 Refining The Linear Model" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Un3ecRdXGC6A" + }, + "source": [ + "You suspected the model was overfitting. This is no real surprise given the number of features you blindly used. It's likely a judicious subset of features would generalize better. `sklearn` has a number of feature selection functions available. The one you'll use here is `SelectKBest` which, as you might guess, selects the k best features. You can read about SelectKBest\n", + "[here](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectKBest.html#sklearn.feature_selection.SelectKBest). `f_regression` is just the [score function](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.f_regression.html#sklearn.feature_selection.f_regression) you're using because you're performing regression. It's important to choose an appropriate one for your machine learning task." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mMLJatIUGC6B" + }, + "source": [ + "### 4.9.1 Define the pipeline" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "hxv4Ir53GC6B" + }, + "source": [ + "Redefine your pipeline to include this feature selection step:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "XQ88OJX9GC6B" + }, + "outputs": [], + "source": [ + "#Code task 16#\n", + "#Add `SelectKBest` as a step in the pipeline between `StandardScaler()` and `LinearRegression()`\n", + "#Don't forget to tell it to use `f_regression` as its score function\n", + "from sklearn.feature_selection import SelectKBest, f_regression\n", + "from sklearn.pipeline import make_pipeline\n", + "from sklearn.impute import SimpleImputer\n", + "from sklearn.preprocessing import StandardScaler, OneHotEncoder # Import OneHotEncoder\n", + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.compose import ColumnTransformer # Import ColumnTransformer\n", + "\n", + "# Identify numerical and categorical features\n", + "numerical_features = X_train.select_dtypes(include=['number']).columns\n", + "categorical_features = X_train.select_dtypes(exclude=['number']).columns\n", + "\n", + "# Create transformers for numerical and categorical features\n", + "numerical_transformer = make_pipeline(\n", + " SimpleImputer(strategy='median'),\n", + " StandardScaler()\n", + ")\n", + "\n", + "categorical_transformer = make_pipeline(\n", + " SimpleImputer(strategy='most_frequent'), # Use most_frequent for categorical features\n", + " OneHotEncoder(handle_unknown='ignore') # One-hot encode categorical features\n", + ")\n", + "\n", + "# Combine transformers using ColumnTransformer\n", + "preprocessor = ColumnTransformer(\n", + " transformers=[\n", + " ('num', numerical_transformer, numerical_features),\n", + " ('cat', categorical_transformer, categorical_features)\n", + " ]\n", + ")\n", + "\n", + "# Create the final pipeline\n", + "pipe = make_pipeline(\n", + " preprocessor,\n", + " SelectKBest(score_func=f_regression), # Use SelectKBest with f_regression as the scoring function\n", + " LinearRegression()\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "qb-CNIVHGC6B" + }, + "source": [ + "### 4.9.2 Fit the pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "svCBhQ59GC6B", + "outputId": "6937c6f7-d1ca-41c6-c99b-1c5ae27d41d5", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 262 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Pipeline(steps=[('columntransformer',\n", + " ColumnTransformer(transformers=[('num',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='median')),\n", + " ('standardscaler',\n", + " StandardScaler())]),\n", + " Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n", + " 'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n", + " 'total_chairs', 'Runs', 'TerrainParks', 'Lon...\n", + " 'averageSnowfall', 'AdultWeekday', 'projectedDaysOpen',\n", + " 'NightSkiing_ac'],\n", + " dtype='object')),\n", + " ('cat',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='most_frequent')),\n", + " ('onehotencoder',\n", + " OneHotEncoder(handle_unknown='ignore'))]),\n", + " Index(['Name', 'Region', 'state'], dtype='object'))])),\n", + " ('selectkbest',\n", + " SelectKBest(score_func=)),\n", + " ('linearregression', LinearRegression())])" + ], + "text/html": [ + "
Pipeline(steps=[('columntransformer',\n",
+              "                 ColumnTransformer(transformers=[('num',\n",
+              "                                                  Pipeline(steps=[('simpleimputer',\n",
+              "                                                                   SimpleImputer(strategy='median')),\n",
+              "                                                                  ('standardscaler',\n",
+              "                                                                   StandardScaler())]),\n",
+              "                                                  Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n",
+              "       'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n",
+              "       'total_chairs', 'Runs', 'TerrainParks', 'Lon...\n",
+              "       'averageSnowfall', 'AdultWeekday', 'projectedDaysOpen',\n",
+              "       'NightSkiing_ac'],\n",
+              "      dtype='object')),\n",
+              "                                                 ('cat',\n",
+              "                                                  Pipeline(steps=[('simpleimputer',\n",
+              "                                                                   SimpleImputer(strategy='most_frequent')),\n",
+              "                                                                  ('onehotencoder',\n",
+              "                                                                   OneHotEncoder(handle_unknown='ignore'))]),\n",
+              "                                                  Index(['Name', 'Region', 'state'], dtype='object'))])),\n",
+              "                ('selectkbest',\n",
+              "                 SelectKBest(score_func=<function f_regression at 0x7a49d3039ab0>)),\n",
+              "                ('linearregression', LinearRegression())])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 138 + } + ], + "source": [ + "pipe.fit(X_train, y_train_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fMJISkvYGC6B" + }, + "source": [ + "### 4.9.3 Assess performance on the train and test set" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "T6Je_FN4GC6B" + }, + "outputs": [], + "source": [ + "y_tr_pred = pipe.predict(X_train)\n", + "y_te_pred = pipe.predict(X_test)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "WicXqbcCGC6C", + "outputId": "ab1527c4-be50-4ae0-dd66-25575bc52a68", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.8957999130320693, 0.8383615057013177)" + ] + }, + "metadata": {}, + "execution_count": 140 + } + ], + "source": [ + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "w_OV3kTDGC6C", + "outputId": "5fb5e706-f77e-488c-fcb2-c09b4de20b3b", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(6.208308232381622, 5.912436455224302)" + ] + }, + "metadata": {}, + "execution_count": 141 + } + ], + "source": [ + "mean_absolute_error(y_train_imputed, y_tr_pred), mean_absolute_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "65YIuKJJGC6C" + }, + "source": [ + "This has made things worse! Clearly selecting a subset of features has an impact on performance. `SelectKBest` defaults to k=10. You've just seen that 10 is worse than using all features. What is the best k? You could create a new pipeline with a different value of k:" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "P740pWUvGC6C" + }, + "source": [ + "### 4.9.4 Define a new pipeline to select a different number of features" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "KehdI6w8GC6C" + }, + "outputs": [], + "source": [ + "# Create a transformer for numerical features\n", + "numerical_transformer = make_pipeline(\n", + " SimpleImputer(strategy='median'),\n", + " StandardScaler()\n", + ")\n", + "\n", + "# Create a transformer for categorical features\n", + "categorical_transformer = make_pipeline(\n", + " SimpleImputer(strategy='most_frequent'), # Use a suitable strategy for categorical data\n", + " OneHotEncoder(handle_unknown='ignore') # Handle unknown categories during testing\n", + ")\n", + "\n", + "# Combine transformers using ColumnTransformer\n", + "# Explicitly specify numerical features instead of relying on exclusion\n", + "numerical_features = X_train.select_dtypes(include=['float', 'int']).columns # Dynamically select numerical columns\n", + "preprocessor = ColumnTransformer(\n", + " transformers=[\n", + " ('num', numerical_transformer, numerical_features), # Pass the list of numerical features\n", + " ('cat', categorical_transformer, categorical_features)\n", + " ]\n", + ")\n", + "\n", + "# Update the pipeline with the preprocessor\n", + "pipe15 = make_pipeline(\n", + " preprocessor,\n", + " SelectKBest(f_regression, k=15),\n", + " LinearRegression()\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Y5dNWCkeGC6C" + }, + "source": [ + "### 4.9.5 Fit the pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "a5JNDYW_GC6C", + "outputId": "25f25ece-fff3-4b12-c476-223d80848202", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 262 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Pipeline(steps=[('columntransformer',\n", + " ColumnTransformer(transformers=[('num',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='median')),\n", + " ('standardscaler',\n", + " StandardScaler())]),\n", + " Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n", + " 'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n", + " 'total_chairs', 'Runs', 'TerrainParks', 'Lon...\n", + " 'averageSnowfall', 'AdultWeekday', 'projectedDaysOpen',\n", + " 'NightSkiing_ac'],\n", + " dtype='object')),\n", + " ('cat',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='most_frequent')),\n", + " ('onehotencoder',\n", + " OneHotEncoder(handle_unknown='ignore'))]),\n", + " ['Name'])])),\n", + " ('selectkbest',\n", + " SelectKBest(k=15,\n", + " score_func=)),\n", + " ('linearregression', LinearRegression())])" + ], + "text/html": [ + "
Pipeline(steps=[('columntransformer',\n",
+              "                 ColumnTransformer(transformers=[('num',\n",
+              "                                                  Pipeline(steps=[('simpleimputer',\n",
+              "                                                                   SimpleImputer(strategy='median')),\n",
+              "                                                                  ('standardscaler',\n",
+              "                                                                   StandardScaler())]),\n",
+              "                                                  Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n",
+              "       'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n",
+              "       'total_chairs', 'Runs', 'TerrainParks', 'Lon...\n",
+              "       'averageSnowfall', 'AdultWeekday', 'projectedDaysOpen',\n",
+              "       'NightSkiing_ac'],\n",
+              "      dtype='object')),\n",
+              "                                                 ('cat',\n",
+              "                                                  Pipeline(steps=[('simpleimputer',\n",
+              "                                                                   SimpleImputer(strategy='most_frequent')),\n",
+              "                                                                  ('onehotencoder',\n",
+              "                                                                   OneHotEncoder(handle_unknown='ignore'))]),\n",
+              "                                                  ['Name'])])),\n",
+              "                ('selectkbest',\n",
+              "                 SelectKBest(k=15,\n",
+              "                             score_func=<function f_regression at 0x7a49d3039ab0>)),\n",
+              "                ('linearregression', LinearRegression())])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 154 + } + ], + "source": [ + "pipe15.fit(X_train, y_train_imputed)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "9iJedhR4GC6D" + }, + "source": [ + "### 4.9.6 Assess performance on train and test data" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "BIukKDDnGC6D" + }, + "outputs": [], + "source": [ + "y_tr_pred = pipe15.predict(X_train)\n", + "y_te_pred = pipe15.predict(X_test)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "2aF_4Hp-GC6D", + "outputId": "65fc1b66-82f5-46f7-aa14-1f4a82566d93", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.9020218697734047, 0.8357076120278937)" + ] + }, + "metadata": {}, + "execution_count": 157 + } + ], + "source": [ + "r2_score(y_train_imputed, y_tr_pred), r2_score(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "rjdpOxR3GC6D", + "outputId": "608e478a-7d07-486d-e687-303a6c93d7ce", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(6.016721660114301, 5.924525833393587)" + ] + }, + "metadata": {}, + "execution_count": 158 + } + ], + "source": [ + "mean_absolute_error(y_train_imputed, y_tr_pred), mean_absolute_error(y_test_imputed, y_te_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "TzB0p9PqGC6D" + }, + "source": [ + "You could keep going, trying different values of k, training a model, measuring performance on the test set, and then picking the model with the best test set performance. There's a fundamental problem with this approach: _you're tuning the model to the arbitrary test set_! If you continue this way you'll end up with a model works well on the particular quirks of our test set _but fails to generalize to new data_. The whole point of keeping a test set is for it to be a set of that new data, to check how well our model might perform on data it hasn't seen.\n", + "\n", + "The way around this is a technique called _cross-validation_. You partition the training set into k folds, train our model on k-1 of those folds, and calculate performance on the fold not used in training. This procedure then cycles through k times with a different fold held back each time. Thus you end up building k models on k sets of data with k estimates of how the model performs on unseen data but without having to touch the test set." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "BLJNDobLGC6D" + }, + "source": [ + "### 4.9.7 Assessing performance using cross-validation" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "wKpJO_EHGC6D" + }, + "outputs": [], + "source": [ + "cv_results = cross_validate(pipe15, X_train, y_train_imputed, cv=5)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "67HH6DYcGC6E", + "outputId": "5d35ce26-83ee-456d-ba2b-eb401d3d68ba", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([0.86636716, 0.86388936, 0.83936053, 0.93953111, 0.85081919])" + ] + }, + "metadata": {}, + "execution_count": 160 + } + ], + "source": [ + "cv_scores = cv_results['test_score']\n", + "cv_scores" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QPxe3b0tGC6E" + }, + "source": [ + "Without using the same random state for initializing the CV folds, your actual numbers will be different." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "0R56F1U4GC6E", + "outputId": "58b880dd-acc2-43ce-d41b-61956f2ffe57", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.8719934714224109, 0.03513381019464692)" + ] + }, + "metadata": {}, + "execution_count": 161 + } + ], + "source": [ + "np.mean(cv_scores), np.std(cv_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5yTYSRO3GC6E" + }, + "source": [ + "These results highlight that assessing model performance in inherently open to variability. You'll get different results depending on the quirks of which points are in which fold. An advantage of this is that you can also obtain an estimate of the variability, or uncertainty, in your performance estimate." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "weupfQrqGC6E", + "outputId": "04d2d147-bfee-4ffd-bb1b-b22ad5b82efc", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([0.8 , 0.94])" + ] + }, + "metadata": {}, + "execution_count": 162 + } + ], + "source": [ + "np.round((np.mean(cv_scores) - 2 * np.std(cv_scores), np.mean(cv_scores) + 2 * np.std(cv_scores)), 2)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "JE9rhbdjGC6E" + }, + "source": [ + "### 4.9.8 Hyperparameter search using GridSearchCV" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4xFSt1NgGC6F" + }, + "source": [ + "Pulling the above together, we have:\n", + "* a pipeline that\n", + " * imputes missing values\n", + " * scales the data\n", + " * selects the k best features\n", + " * trains a linear regression model\n", + "* a technique (cross-validation) for estimating model performance\n", + "\n", + "Now you want to use cross-validation for multiple values of k and use cross-validation to pick the value of k that gives the best performance. `make_pipeline` automatically names each step as the lowercase name of the step and the parameters of the step are then accessed by appending a double underscore followed by the parameter name. You know the name of the step will be 'selectkbest' and you know the parameter is 'k'.\n", + "\n", + "You can also list the names of all the parameters in a pipeline like this:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "5K3LrM7QGC6F", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "8d9c1a69-c463-4507-e0bf-692e1d525981" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "dict_keys(['memory', 'steps', 'verbose', 'columntransformer', 'selectkbest', 'linearregression', 'columntransformer__n_jobs', 'columntransformer__remainder', 'columntransformer__sparse_threshold', 'columntransformer__transformer_weights', 'columntransformer__transformers', 'columntransformer__verbose', 'columntransformer__verbose_feature_names_out', 'columntransformer__num', 'columntransformer__cat', 'columntransformer__num__memory', 'columntransformer__num__steps', 'columntransformer__num__verbose', 'columntransformer__num__simpleimputer', 'columntransformer__num__standardscaler', 'columntransformer__num__simpleimputer__add_indicator', 'columntransformer__num__simpleimputer__copy', 'columntransformer__num__simpleimputer__fill_value', 'columntransformer__num__simpleimputer__keep_empty_features', 'columntransformer__num__simpleimputer__missing_values', 'columntransformer__num__simpleimputer__strategy', 'columntransformer__num__simpleimputer__verbose', 'columntransformer__num__standardscaler__copy', 'columntransformer__num__standardscaler__with_mean', 'columntransformer__num__standardscaler__with_std', 'columntransformer__cat__memory', 'columntransformer__cat__steps', 'columntransformer__cat__verbose', 'columntransformer__cat__simpleimputer', 'columntransformer__cat__onehotencoder', 'columntransformer__cat__simpleimputer__add_indicator', 'columntransformer__cat__simpleimputer__copy', 'columntransformer__cat__simpleimputer__fill_value', 'columntransformer__cat__simpleimputer__keep_empty_features', 'columntransformer__cat__simpleimputer__missing_values', 'columntransformer__cat__simpleimputer__strategy', 'columntransformer__cat__simpleimputer__verbose', 'columntransformer__cat__onehotencoder__categories', 'columntransformer__cat__onehotencoder__drop', 'columntransformer__cat__onehotencoder__dtype', 'columntransformer__cat__onehotencoder__handle_unknown', 'columntransformer__cat__onehotencoder__max_categories', 'columntransformer__cat__onehotencoder__min_frequency', 'columntransformer__cat__onehotencoder__sparse', 'columntransformer__cat__onehotencoder__sparse_output', 'selectkbest__k', 'selectkbest__score_func', 'linearregression__copy_X', 'linearregression__fit_intercept', 'linearregression__n_jobs', 'linearregression__positive'])" + ] + }, + "metadata": {}, + "execution_count": 163 + } + ], + "source": [ + "#Code task 18#\n", + "#Call `pipe`'s `get_params()` method to get a dict of available parameters and print their names\n", + "#using dict's `keys()` method\n", + "pipe.get_params().keys()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xszf93zgGC6F" + }, + "source": [ + "The above can be particularly useful as your pipelines becomes more complex (you can even nest pipelines within pipelines)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "9VGINeR5GC6F" + }, + "outputs": [], + "source": [ + "k = [k+1 for k in range(len(X_train.columns))]\n", + "grid_params = {'selectkbest__k': k}" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PPkUIh_rGC6F" + }, + "source": [ + "Now you have a range of `k` to investigate. Is 1 feature best? 2? 3? 4? All of them? You could write a for loop and iterate over each possible value, doing all the housekeeping oyurselves to track the best value of k. But this is a common task so there's a built in function in `sklearn`. This is [`GridSearchCV`](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html).\n", + "This takes the pipeline object, in fact it takes anything with a `.fit()` and `.predict()` method. In simple cases with no feature selection or imputation or feature scaling etc. you may see the classifier or regressor object itself directly passed into `GridSearchCV`. The other key input is the parameters and values to search over. Optional parameters include the cross-validation strategy and number of CPUs to use." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "dpO48JYDGC6F" + }, + "outputs": [], + "source": [ + "lr_grid_cv = GridSearchCV(pipe, param_grid=grid_params, cv=5, n_jobs=-1)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "rpVenczFGC6G", + "outputId": "134b3eb3-704a-4906-e462-c144943a2d73", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 288 + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "GridSearchCV(cv=5,\n", + " estimator=Pipeline(steps=[('columntransformer',\n", + " ColumnTransformer(transformers=[('num',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='median')),\n", + " ('standardscaler',\n", + " StandardScaler())]),\n", + " Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n", + " 'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n", + " 'total_chairs...\n", + " SimpleImputer(strategy='most_frequent')),\n", + " ('onehotencoder',\n", + " OneHotEncoder(handle_unknown='ignore'))]),\n", + " Index(['Name', 'Region', 'state'], dtype='object'))])),\n", + " ('selectkbest',\n", + " SelectKBest(score_func=)),\n", + " ('linearregression',\n", + " LinearRegression())]),\n", + " n_jobs=-1,\n", + " param_grid={'selectkbest__k': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,\n", + " 12, 13, 14, 15, 16, 17, 18, 19, 20,\n", + " 21, 22, 23, 24, 25, 26]})" + ], + "text/html": [ + "
GridSearchCV(cv=5,\n",
+              "             estimator=Pipeline(steps=[('columntransformer',\n",
+              "                                        ColumnTransformer(transformers=[('num',\n",
+              "                                                                         Pipeline(steps=[('simpleimputer',\n",
+              "                                                                                          SimpleImputer(strategy='median')),\n",
+              "                                                                                         ('standardscaler',\n",
+              "                                                                                          StandardScaler())]),\n",
+              "                                                                         Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n",
+              "       'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n",
+              "       'total_chairs...\n",
+              "                                                                                          SimpleImputer(strategy='most_frequent')),\n",
+              "                                                                                         ('onehotencoder',\n",
+              "                                                                                          OneHotEncoder(handle_unknown='ignore'))]),\n",
+              "                                                                         Index(['Name', 'Region', 'state'], dtype='object'))])),\n",
+              "                                       ('selectkbest',\n",
+              "                                        SelectKBest(score_func=<function f_regression at 0x7a49d3039ab0>)),\n",
+              "                                       ('linearregression',\n",
+              "                                        LinearRegression())]),\n",
+              "             n_jobs=-1,\n",
+              "             param_grid={'selectkbest__k': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,\n",
+              "                                            12, 13, 14, 15, 16, 17, 18, 19, 20,\n",
+              "                                            21, 22, 23, 24, 25, 26]})
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 166 + } + ], + "source": [ + "lr_grid_cv.fit(X_train, y_train_imputed)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Gm1hUNoQGC6G" + }, + "outputs": [], + "source": [ + "score_mean = lr_grid_cv.cv_results_['mean_test_score']\n", + "score_std = lr_grid_cv.cv_results_['std_test_score']\n", + "cv_k = [k for k in lr_grid_cv.cv_results_['param_selectkbest__k']]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "PGA45-iSGC6G", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "2b60edc8-cd65-4122-d252-8dadd45be062" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "{'selectkbest__k': 11}" + ] + }, + "metadata": {}, + "execution_count": 168 + } + ], + "source": [ + "#Code task 19#\n", + "#Print the `best_params_` attribute of `lr_grid_cv`\n", + "lr_grid_cv.best_params_" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "LZsRd14cGC6G", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 487 + }, + "outputId": "d7d3d716-66a1-4e0d-acd0-2ef572e13230" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "#Code task 20#\n", + "#Assign the value of k from the above dict of `best_params_` and assign it to `best_k`\n", + "best_k = lr_grid_cv.best_params_['selectkbest__k']\n", + "plt.subplots(figsize=(10, 5))\n", + "plt.errorbar(cv_k, score_mean, yerr=score_std)\n", + "plt.axvline(x=best_k, c='r', ls='--', alpha=.5)\n", + "plt.xlabel('k')\n", + "plt.ylabel('CV score (r-squared)')\n", + "plt.title('Pipeline mean CV score (error bars +/- 1sd)');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "gXOKq34-GC6G" + }, + "source": [ + "The above suggests a good value for k is 8. There was an initial rapid increase with k, followed by a slow decline. Also noticeable is the variance of the results greatly increase above k=8. As you increasingly overfit, expect greater swings in performance as different points move in and out of the train/test folds." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "NMhS_i4OGC6G" + }, + "source": [ + "Which features were most useful? Step into your best model, shown below. Starting with the fitted grid search object, you get the best estimator, then the named step 'selectkbest', for which you can its `get_support()` method for a logical mask of the features selected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ZV5D8DlRGC6G" + }, + "outputs": [], + "source": [ + "selected = lr_grid_cv.best_estimator_.named_steps.selectkbest.get_support()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "NQX8ihXDGC6G" + }, + "source": [ + "Similarly, instead of using the 'selectkbest' named step, you can access the named step for the linear regression model and, from that, grab the model coefficients via its `coef_` attribute:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "YJm7iJ2mGC6H", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "0369c15a-4453-4e63-871a-5a404670ef6f" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "AdultWeekday 21.087076\n", + "vertical_drop 3.276549\n", + "total_chairs 2.314675\n", + "daysOpenLastYear 1.490134\n", + "fastQuads 0.630488\n", + "averageSnowfall -0.110770\n", + "LongestRun_mi -0.909929\n", + "SkiableTerrain_ac -0.991550\n", + "Runs -1.098028\n", + "projectedDaysOpen -1.177025\n", + "summit_elev -2.198581\n", + "dtype: float64" + ] + }, + "metadata": {}, + "execution_count": 178 + } + ], + "source": [ + "#Code task 21#\n", + "#Get the linear model coefficients from the `coef_` attribute and store in `coefs`,\n", + "#get the matching feature names from the column names of the dataframe,\n", + "#and display the results as a pandas Series with `coefs` as the values and `features` as the index,\n", + "#sorting the values in descending order\n", + "coefs = lr_grid_cv.best_estimator_.named_steps.linearregression.coef_\n", + "\n", + "# Refit the SelectKBest on the entire training set to get the correct support\n", + "selector = lr_grid_cv.best_estimator_.named_steps.selectkbest\n", + "\n", + "# Handle non-numeric columns (example: one-hot encoding)\n", + "X_train_encoded = pd.get_dummies(X_train, columns=[col for col in X_train.columns if X_train[col].dtype == 'object'])\n", + "\n", + "# Fill missing values (NaNs) with a suitable strategy, e.g., the mean\n", + "X_train_encoded.fillna(X_train_encoded.mean(), inplace=True) # Fill NaNs with mean\n", + "\n", + "# Fit the selector after encoding and filling NaNs\n", + "selector.fit(X_train_encoded, y_train_imputed)\n", + "\n", + "# Get feature names from the encoded DataFrame\n", + "features = X_train_encoded.columns[selector.get_support()]\n", + "\n", + "pd.Series(coefs, index=features).sort_values(ascending=False) # Use sort_values instead of sort" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "RTcVJbkoGC6H" + }, + "source": [ + "These results suggest that vertical drop is your biggest positive feature. This makes intuitive sense and is consistent with what you saw during the EDA work. Also, you see the area covered by snow making equipment is a strong positive as well. People like guaranteed skiing! The skiable terrain area is negatively associated with ticket price! This seems odd. People will pay less for larger resorts? There could be all manner of reasons for this. It could be an effect whereby larger resorts can host more visitors at any one time and so can charge less per ticket. As has been mentioned previously, the data are missing information about visitor numbers. Bear in mind, the coefficient for skiable terrain is negative _for this model_. For example, if you kept the total number of chairs and fastQuads constant, but increased the skiable terrain extent, you might imagine the resort is worse off because the chairlift capacity is stretched thinner." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "NmXpSWQBGC6H" + }, + "source": [ + "## 4.10 Random Forest Model" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "xYUq8iZaGC6H" + }, + "source": [ + "A model that can work very well in a lot of cases is the random forest. For regression, this is provided by `sklearn`'s `RandomForestRegressor` class.\n", + "\n", + "Time to stop the bad practice of repeatedly checking performance on the test split. Instead, go straight from defining the pipeline to assessing performance using cross-validation. `cross_validate` will perform the fitting as part of the process. This uses the default settings for the random forest so you'll then proceed to investigate some different hyperparameters." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "kVzdHMacGC6H" + }, + "source": [ + "### 4.10.1 Define the pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "JEQ7qTtpGC6H" + }, + "outputs": [], + "source": [ + "#Code task 22#\n", + "#Define a pipeline comprising the steps:\n", + "#SimpleImputer() with a strategy of 'median'\n", + "#StandardScaler(),\n", + "#and then RandomForestRegressor() with a random state of 47\n", + "RF_pipe = make_pipeline(\n", + " SimpleImputer(strategy=\"median\"),\n", + " StandardScaler(),\n", + " RandomForestRegressor(random_state=47)\n", + ")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QNV7aZWRGC6H" + }, + "source": [ + "### 4.10.2 Fit and assess performance using cross-validation" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "YuQpZKbKGC6I" + }, + "outputs": [], + "source": [ + "#Code task 23#\n", + "#Call `cross_validate` to estimate the pipeline's performance.\n", + "#Pass it the random forest pipe object, `X_train` and `y_train`,\n", + "#and get it to use 5-fold cross-validation\n", + "rf_default_cv_results = cross_validate(RF_pipe, X_train_encoded, y_train_imputed, cv=5) # Changed cv to 5" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "XMu9K7yxGC6I", + "outputId": "b0570728-6a01-42e1-cbea-bfa29fcab422", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([0.85578693, 0.84500978, 0.85536643, 0.86738887, 0.84526525])" + ] + }, + "metadata": {}, + "execution_count": 184 + } + ], + "source": [ + "rf_cv_scores = rf_default_cv_results['test_score']\n", + "rf_cv_scores" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "D-Z9Y-mEGC6I", + "outputId": "0d3b823d-2ef1-482f-898e-b54489fbe2ec", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.8537634529648812, 0.008260294127408406)" + ] + }, + "metadata": {}, + "execution_count": 185 + } + ], + "source": [ + "np.mean(rf_cv_scores), np.std(rf_cv_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "PfG_kTu6GC6I" + }, + "source": [ + "### 4.10.3 Hyperparameter search using GridSearchCV" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Ou69-GkCGC6I" + }, + "source": [ + "Random forest has a number of hyperparameters that can be explored, however here you'll limit yourselves to exploring some different values for the number of trees. You'll try it with and without feature scaling, and try both the mean and median as strategies for imputing missing values." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "lC2t7aY1GC6I", + "outputId": "66a9b626-0cd9-47e3-e7db-13d330c9988a", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "{'randomforestregressor__n_estimators': [10,\n", + " 12,\n", + " 16,\n", + " 20,\n", + " 26,\n", + " 33,\n", + " 42,\n", + " 54,\n", + " 69,\n", + " 88,\n", + " 112,\n", + " 143,\n", + " 183,\n", + " 233,\n", + " 297,\n", + " 379,\n", + " 483,\n", + " 615,\n", + " 784,\n", + " 1000],\n", + " 'standardscaler': [StandardScaler(), None],\n", + " 'simpleimputer__strategy': ['mean', 'median']}" + ] + }, + "metadata": {}, + "execution_count": 186 + } + ], + "source": [ + "n_est = [int(n) for n in np.logspace(start=1, stop=3, num=20)]\n", + "grid_params = {\n", + " 'randomforestregressor__n_estimators': n_est,\n", + " 'standardscaler': [StandardScaler(), None],\n", + " 'simpleimputer__strategy': ['mean', 'median']\n", + "}\n", + "grid_params" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "p1Pcry71GC6I" + }, + "outputs": [], + "source": [ + "#Code task 24#\n", + "#Call `GridSearchCV` with the random forest pipeline, passing in the above `grid_params`\n", + "#dict for parameters to evaluate, 5-fold cross-validation, and all available CPU cores (if desired)\n", + "rf_grid_cv = GridSearchCV(RF_pipe, param_grid=grid_params, cv=5, n_jobs=-1)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "m8FCP8UVGC6J", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 187 + }, + "outputId": "9ca3edeb-973e-41e1-f8d3-1e01eab262c4" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "GridSearchCV(cv=5,\n", + " estimator=Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='median')),\n", + " ('standardscaler', StandardScaler()),\n", + " ('randomforestregressor',\n", + " RandomForestRegressor(random_state=47))]),\n", + " n_jobs=-1,\n", + " param_grid={'randomforestregressor__n_estimators': [10, 12, 16, 20,\n", + " 26, 33, 42, 54,\n", + " 69, 88, 112,\n", + " 143, 183, 233,\n", + " 297, 379, 483,\n", + " 615, 784,\n", + " 1000],\n", + " 'simpleimputer__strategy': ['mean', 'median'],\n", + " 'standardscaler': [StandardScaler(), None]})" + ], + "text/html": [ + "
GridSearchCV(cv=5,\n",
+              "             estimator=Pipeline(steps=[('simpleimputer',\n",
+              "                                        SimpleImputer(strategy='median')),\n",
+              "                                       ('standardscaler', StandardScaler()),\n",
+              "                                       ('randomforestregressor',\n",
+              "                                        RandomForestRegressor(random_state=47))]),\n",
+              "             n_jobs=-1,\n",
+              "             param_grid={'randomforestregressor__n_estimators': [10, 12, 16, 20,\n",
+              "                                                                 26, 33, 42, 54,\n",
+              "                                                                 69, 88, 112,\n",
+              "                                                                 143, 183, 233,\n",
+              "                                                                 297, 379, 483,\n",
+              "                                                                 615, 784,\n",
+              "                                                                 1000],\n",
+              "                         'simpleimputer__strategy': ['mean', 'median'],\n",
+              "                         'standardscaler': [StandardScaler(), None]})
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 192 + } + ], + "source": [ + "#Code task 25#\n", + "#Now call the `GridSearchCV`'s `fit()` method with `X_train` and `y_train` as arguments\n", + "#to actually start the grid search. This may take a minute or two.\n", + "rf_grid_cv.fit(X_train_encoded, y_train_imputed)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Sz1-5h0ZGC6J", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "702db723-c309-4b34-e0ec-334daf225cec" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "{'randomforestregressor__n_estimators': 379,\n", + " 'simpleimputer__strategy': 'mean',\n", + " 'standardscaler': None}" + ] + }, + "metadata": {}, + "execution_count": 193 + } + ], + "source": [ + "#Code task 26#\n", + "#Print the best params (`best_params_` attribute) from the grid search\n", + "rf_grid_cv.best_params_" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "5HEzyZVXGC6J" + }, + "source": [ + "It looks like imputing with the median helps, but scaling the features doesn't." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "AFzbGSHhGC6J", + "outputId": "46823dac-23bc-4760-86d1-6f53be447e89", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([0.85269617, 0.84469567, 0.8611613 , 0.87258701, 0.84291051])" + ] + }, + "metadata": {}, + "execution_count": 194 + } + ], + "source": [ + "rf_best_cv_results = cross_validate(rf_grid_cv.best_estimator_, X_train_encoded, y_train_imputed, cv=5)\n", + "rf_best_scores = rf_best_cv_results['test_score']\n", + "rf_best_scores" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "ZOgXZIMLGC6J", + "outputId": "2ebac412-2be6-4aea-ea17-58ea0194645e", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(0.8548101311774511, 0.010997518424168197)" + ] + }, + "metadata": {}, + "execution_count": 195 + } + ], + "source": [ + "np.mean(rf_best_scores), np.std(rf_best_scores)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "FF35kRwnGC6K" + }, + "source": [ + "You've marginally improved upon the default CV results. Random forest has many more hyperparameters you could tune, but we won't dive into that here." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "whQ7rhwPGC6K", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 780 + }, + "outputId": "2658d67c-cdc0-4aa0-cb17-a0db5397c8ee" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "#Code task 27#\n", + "#Plot a barplot of the random forest's feature importances,\n", + "#assigning the `feature_importances_` attribute of\n", + "#`rf_grid_cv.best_estimator_.named_steps.randomforestregressor` to the name `imps` to then\n", + "#create a pandas Series object of the feature importances, with the index given by the\n", + "#training data column names, sorting the values in descending order\n", + "plt.subplots(figsize=(10, 5))\n", + "imps = rf_grid_cv.best_estimator_.named_steps.randomforestregressor.feature_importances_\n", + "rf_feat_imps = pd.Series(data=imps, index=X_train_encoded.columns).sort_values(ascending=False)\n", + "rf_feat_imps.plot(kind='bar')\n", + "plt.xlabel('features')\n", + "plt.ylabel('importance')\n", + "plt.title('Best random forest regressor feature importances');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "r6yvTzzfGC6K" + }, + "source": [ + "Encouragingly, the dominant top four features are in common with your linear model:\n", + "* fastQuads\n", + "* Runs\n", + "* Snow Making_ac\n", + "* vertical_drop" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UIOUaNT3GC6K" + }, + "source": [ + "## 4.11 Final Model Selection" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Ipn7j1bZGC6K" + }, + "source": [ + "Time to select your final model to use for further business modeling! It would be good to revisit the above model selection; there is undoubtedly more that could be done to explore possible hyperparameters.\n", + "It would also be worthwhile to investigate removing the least useful features. Gathering or calculating, and storing, features adds business cost and dependencies, so if features genuinely are not needed they should be removed.\n", + "Building a simpler model with fewer features can also have the advantage of being easier to sell (and/or explain) to stakeholders.\n", + "Certainly there seem to be four strong features here and so a model using only those would probably work well.\n", + "However, you want to explore some different scenarios where other features vary so keep the fuller\n", + "model for now.\n", + "The business is waiting for this model and you have something that you have confidence in to be much better than guessing with the average price.\n", + "\n", + "Or, rather, you have two \"somethings\". You built a best linear model and a best random forest model. You need to finally choose between them. You can calculate the mean absolute error using cross-validation. Although `cross-validate` defaults to the $R^2$ [metric for scoring](https://scikit-learn.org/stable/modules/model_evaluation.html#scoring) regression, you can specify the mean absolute error as an alternative via\n", + "the `scoring` parameter." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "RNvcKTrFGC6K" + }, + "source": [ + "### 4.11.1 Linear regression model performance" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Q0EpP1ZkGC6K", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "ba8e38bc-b1f3-4b1a-ce20-63d82cc38ed1" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "ColumnTransformer(transformers=[('num',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='median')),\n", + " ('standardscaler',\n", + " StandardScaler())]),\n", + " Index(['summit_elev', 'vertical_drop', 'base_elev', 'trams', 'fastEight',\n", + " 'fastSixes', 'fastQuads', 'quad', 'triple', 'double', 'surface',\n", + " 'total_chairs', 'Runs', 'TerrainParks', 'LongestRun_mi',\n", + " 'SkiableTerrain_ac', 'Snow Making_ac', 'daysOpenLastYear', 'yearsOpen',\n", + " 'averageSnowfall', 'AdultWeekday', 'projectedDaysOpen',\n", + " 'NightSkiing_ac'],\n", + " dtype='object')),\n", + " ('cat',\n", + " Pipeline(steps=[('simpleimputer',\n", + " SimpleImputer(strategy='most_frequent')),\n", + " ('onehotencoder',\n", + " OneHotEncoder(handle_unknown='ignore'))]),\n", + " Index(['Name', 'Region', 'state'], dtype='object'))])\n" + ] + } + ], + "source": [ + "# Get the name of the ColumnTransformer step\n", + "lr_neg_mae = lr_grid_cv.best_estimator_.named_steps.keys()\n", + "\n", + "# Print the ColumnTransformer\n", + "print(lr_grid_cv.best_estimator_.named_steps[list(column_transformer_name)[0]])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "1FjJ4V8dGC6K", + "outputId": "5c1d9f33-fe79-473e-a92b-6e55fc5293a0", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(-0.8676236539211758, 0.0033392212546386312)" + ] + }, + "metadata": {}, + "execution_count": 212 + } + ], + "source": [ + "# Get the cross-validation results for the best estimator\n", + "lr_cv_results = lr_grid_cv.cv_results_\n", + "\n", + "# Calculate mean and standard deviation of the negative MAE scores\n", + "lr_mae_mean = np.mean(-1 * lr_cv_results['mean_test_score']) # Use 'mean_test_score' from cv_results_\n", + "lr_mae_std = np.std(-1 * lr_cv_results['mean_test_score'])\n", + "\n", + "# Print the results\n", + "lr_mae_mean, lr_mae_std" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "lHvJdY8RGC6L", + "outputId": "5da5f478-2d78-458a-81ec-8f32df77b000", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/sklearn/base.py:439: UserWarning: X does not have valid feature names, but SelectKBest was fitted with feature names\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "5.7810516759903505" + ] + }, + "metadata": {}, + "execution_count": 216 + } + ], + "source": [ + "mean_absolute_error(y_test_imputed, lr_grid_cv.best_estimator_.predict(X_test))" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "G1GgDYcsGC6L" + }, + "source": [ + "### 4.11.2 Random forest regression model performance" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "SvQ4-LYVGC6L" + }, + "outputs": [], + "source": [ + "rf_neg_mae = cross_validate(rf_grid_cv.best_estimator_, X_train_encoded, y_train_imputed,\n", + " scoring='neg_mean_absolute_error', cv=5, n_jobs=-1)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "-BfR-qkfGC6L", + "outputId": "ef263cf2-714a-40ca-f605-5543d38fd9ec", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(6.29875820384154, 0.5826861478214918)" + ] + }, + "metadata": {}, + "execution_count": 218 + } + ], + "source": [ + "rf_mae_mean = np.mean(-1 * rf_neg_mae['test_score'])\n", + "rf_mae_std = np.std(-1 * rf_neg_mae['test_score'])\n", + "rf_mae_mean, rf_mae_std" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "LBmVdE7yGC6L", + "outputId": "dbd84fe6-83be-4020-98d2-8cd579552924", + "colab": { + "base_uri": "https://localhost:8080/" + } + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Index(['Name', 'Region', 'state', 'summit_elev', 'vertical_drop', 'base_elev',\n", + " 'trams', 'fastEight', 'fastSixes', 'fastQuads', 'quad', 'triple',\n", + " 'double', 'surface', 'total_chairs', 'Runs', 'TerrainParks',\n", + " 'LongestRun_mi', 'SkiableTerrain_ac', 'Snow Making_ac',\n", + " 'daysOpenLastYear', 'yearsOpen', 'averageSnowfall', 'AdultWeekday',\n", + " 'projectedDaysOpen', 'NightSkiing_ac'],\n", + " dtype='object')\n", + "Error: The following features are missing from X_train: {'Ticket', 'Embarked', 'Cabin', 'Sex'}\n" + ] + } + ], + "source": [ + "# Import OneHotEncoder\n", + "from sklearn.preprocessing import OneHotEncoder\n", + "\n", + "# Assuming categorical features are in columns identified by 'categorical_features'\n", + "# Double-check these column names against your X_train DataFrame\n", + "# Make sure these columns ACTUALLY exist in X_train\n", + "categorical_features = ['Name', 'Sex', 'Ticket', 'Cabin', 'Embarked']\n", + "\n", + "# Create and fit the OneHotEncoder\n", + "encoder = OneHotEncoder(handle_unknown='ignore')\n", + "\n", + "# Verify that X_train contains the specified categorical features\n", + "print(X_train.columns) # Print columns of X_train to check for presence of categorical features\n", + "\n", + "# Check if all categorical features are in X_train\n", + "missing_features = set(categorical_features) - set(X_train.columns)\n", + "if missing_features:\n", + " print(f\"Error: The following features are missing from X_train: {missing_features}\")\n", + "else:\n", + " # Fit on training data using all categorical features\n", + " encoder.fit(X_train[categorical_features])\n", + "\n", + "# ... rest of the code ..." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Mueogux7GC6L" + }, + "source": [ + "### 4.11.3 Conclusion" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "7V9BqNt7GC6M" + }, + "source": [ + "The random forest model has a lower cross-validation mean absolute error by almost \\\\$1. It also exhibits less variability. Verifying performance on the test set produces performance consistent with the cross-validation results." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "w5Ro-D5zGC6M" + }, + "source": [ + "## 4.12 Data quantity assessment" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "iTD6ejcCGC6M" + }, + "source": [ + "Finally, you need to advise the business whether it needs to undertake further data collection. Would more data be useful? We're often led to believe more data is always good, but gathering data invariably has a cost associated with it. Assess this trade off by seeing how performance varies with differing data set sizes. The `learning_curve` function does this conveniently." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Hi1lWLLPGC6M" + }, + "outputs": [], + "source": [ + "fractions = [.2, .25, .3, .35, .4, .45, .5, .6, .75, .8, 1.0]\n", + "train_size, train_scores, test_scores = learning_curve(pipe, X_train, y_train, train_sizes=fractions)\n", + "train_scores_mean = np.mean(train_scores, axis=1)\n", + "train_scores_std = np.std(train_scores, axis=1)\n", + "test_scores_mean = np.mean(test_scores, axis=1)\n", + "test_scores_std = np.std(test_scores, axis=1)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "Zr9kZCNrGC6M", + "outputId": "6cb92f53-4b1e-4f2b-8393-2ccf0b142bbd", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 523 + } + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/matplotlib/axes/_base.py:2503: UserWarning: Warning: converting a masked element to nan.\n", + " xys = np.asarray(xys)\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2IAAAHWCAYAAAAVazrYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABMgElEQVR4nO3de3zP9f//8ft7Zgc7MrMZY061GVFjmgqxjFSEaCmHj6gcKyk+lUMnoUIhHT6flFKM0idJNOS0xEjOIaewIbY5bmzP3x/99v5628HG9hpzu14ur4vez9fz9Xo9Xu/X67123+v1er5txhgjAAAAAIBlnEq6AAAAAAC40RDEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAIBfTp0+XzWbT3r177W0tWrRQixYtLrvssmXLZLPZtGzZsiKtyWazadSoUUW6Tlw/QkJC1LNnzytatqDn7o0mt895cRs1apRsNptl2wNw7SKIAdDu3bv1xBNPqGbNmnJzc5O3t7fuuOMOTZo0SWfPni3p8m4oCxYsIGxdp1avXq1Ro0YpJSWlpEsptd544w3NmzevpMsAgCJhM8aYki4CQMn5/vvv9dBDD8nV1VXdu3dXvXr1lJGRoZUrV2ru3Lnq2bOnPvzww5Iu03LTp09Xr169tGfPHoWEhEiSMjIyJEkuLi75Lrts2TLdfffdWrp0aaGvQgwYMEBTpkxRbj+az507J2dnZzk7OxdqnbDGW2+9paFDhzqcM0UpPT1dTk5OKlu2bKGXLei5e63z9PRU586dNX369CJZX2Zmps6fPy9XV1fLrlJduHBBFy5ckJubmyXbA3Dt4v/mwA1sz549evjhh1W9enUtWbJElStXts/r37+/du3ape+//z7P5bOyspSRkXHD/EJR0r/E3ijvc25Onz4tDw+Pki6jyFzJZ8fV1fWKt1fS5+61qkyZMipTpoyl2yyJP6aUts8PUFpwayJwAxs3bpxOnTql//znPw4hLFvt2rU1ePBg+2ubzaYBAwboiy++UHh4uFxdXbVw4UJJ0oYNG9S2bVt5e3vL09NTrVq10i+//OKwvvPnz2v06NGqU6eO3Nzc5OfnpzvvvFOLFy+290lKSlKvXr1UtWpVubq6qnLlymrfvn2+z3DMmTNHNptNP//8c455H3zwgWw2mzZv3ixJ+v3339WzZ0/7bZiBgYH617/+pb///vuy71duz9n89ddf6tChgzw8PFSpUiU988wzSk9Pz7HsihUr9NBDD6latWpydXVVcHCwnnnmGYdbP3v27KkpU6ZI+ue9zp6y5faMWEHe9+znYFatWqVnn31W/v7+8vDw0IMPPqijR49edr8Lekx++OEHNW/eXF5eXvL29lbjxo01c+ZMhz5xcXGKiIiQu7u7KlasqEcffVQHDx506NOzZ095enpq9+7duvfee+Xl5aVu3bpJ+ifATJw4UeHh4XJzc1NAQICeeOIJnThx4rL7UdBjf/LkST399NMKCQmRq6urKlWqpHvuuUfr16/Pc92jRo3S0KFDJUk1atSwH7vs9yi/z85bb72lpk2bys/PT+7u7oqIiNCcOXNybOPSZ8QKc1wvPXezn2OcPXu2Xn/9dVWtWlVubm5q1aqVdu3alWPbU6ZMUc2aNeXu7q7IyEitWLGiwM+dLV68WHfeead8fX3l6empm2++Wf/+978d+qSnp2vkyJGqXbu2/fPx/PPPO3yWbDabTp8+rU8//dT+/l7umbn33ntP4eHhKleunMqXL69GjRo5nJOXPiOW/fxWbtPF27qa8zC3Z8Syz4958+apXr16cnV1VXh4uP0cudjBgwfVu3dvBQUFydXVVTVq1NBTTz1lv+qZvU8///yz+vXrp0qVKqlq1ar25X/44Qfddddd8vDwkJeXl9q1a6ctW7Y4bKOoPytr1qxRmzZt5OPjo3Llyql58+ZatWrVFa0LKE24IgbcwL777jvVrFlTTZs2LfAyS5Ys0ezZszVgwABVrFhRISEh2rJli+666y55e3vr+eefV9myZfXBBx+oRYsW+vnnn9WkSRNJ//wCMmbMGD3++OOKjIxUWlqa1q1bp/Xr1+uee+6RJHXq1ElbtmzRwIEDFRISoiNHjmjx4sXav39/nrd7tWvXTp6enpo9e7aaN2/uMG/WrFkKDw9XvXr1JP3zS+Gff/6pXr16KTAwUFu2bNGHH36oLVu26JdffinU7Ulnz55Vq1attH//fg0aNEhBQUGaMWOGlixZkqNvXFyczpw5o6eeekp+fn769ddf9d577+mvv/5SXFycJOmJJ57QoUOHtHjxYs2YMeOy2y/o+55t4MCBKl++vEaOHKm9e/dq4sSJGjBggGbNmpXvdgpyTKZPn65//etfCg8P1/Dhw+Xr66sNGzZo4cKFeuSRR+x9evXqpcaNG2vMmDFKTk7WpEmTtGrVKm3YsEG+vr72bV64cEExMTG688479dZbb6lcuXL29yh7PYMGDdKePXs0efJkbdiwQatWrcr3tr2CHvsnn3xSc+bM0YABA1S3bl39/fffWrlypbZt26bbbrst13V37NhRf/zxh7788ktNmDBBFStWlCT5+/vb++T22ZGkSZMm6YEHHlC3bt2UkZGhr776Sg899JDmz5+vdu3a5XtspCs/rpL05ptvysnJSc8995xSU1M1btw4devWTWvWrLH3ef/99zVgwADdddddeuaZZ7R371516NBB5cuXd/gFPzdbtmzRfffdp1tuuUWvvPKKXF1dtWvXLodfwrOysvTAAw9o5cqV6tu3r8LCwrRp0yZNmDBBf/zxh/2ZsBkzZth/dvTt21eSVKtWrTy3/dFHH2nQoEHq3LmzBg8erHPnzun333/XmjVr7OfkpTp27KjatWs7tCUmJmrixImqVKmSve1qzsO8rFy5Ul9//bX69esnLy8vvfvuu+rUqZP2798vPz8/SdKhQ4cUGRmplJQU9e3bV6GhoTp48KDmzJmjM2fOOFz57Nevn/z9/TVixAidPn3a/h726NFDMTExGjt2rM6cOaP3339fd955pzZs2GA/J4vys7JkyRK1bdtWERERGjlypJycnPTJJ5+oZcuWWrFihSIjIwu8LqDUMQBuSKmpqUaSad++fYGXkWScnJzMli1bHNo7dOhgXFxczO7du+1thw4dMl5eXqZZs2b2tgYNGph27drluf4TJ04YSWb8+PEF35H/LzY21lSqVMlcuHDB3nb48GHj5ORkXnnlFXvbmTNnciz75ZdfGklm+fLl9rZPPvnESDJ79uyxtzVv3tw0b97c/nrixIlGkpk9e7a97fTp06Z27dpGklm6dGm+2x0zZoyx2Wxm37599rb+/fubvH40SzIjR460vy7o+569L9HR0SYrK8ve/swzz5gyZcqYlJSUXLdnTMGOSUpKivHy8jJNmjQxZ8+edZiXvb2MjAxTqVIlU69ePYc+8+fPN5LMiBEj7G09evQwksywYcMc1rVixQojyXzxxRcO7QsXLsy1/VIFPfY+Pj6mf//++a4rN+PHj89xzmTL67OTW10ZGRmmXr16pmXLlg7t1atXNz169LC/LsxxvfTcXbp0qZFkwsLCTHp6ur190qRJRpLZtGmTMcaY9PR04+fnZxo3bmzOnz9v7zd9+nQjyWGduZkwYYKRZI4ePZpnnxkzZhgnJyezYsUKh/Zp06YZSWbVqlX2Ng8PD4f3ID/t27c34eHh+fbJ7XN+saNHj5pq1aqZ+vXrm1OnThljrv48HDlyZI7PuCTj4uJidu3aZW/buHGjkWTee+89e1v37t2Nk5OTWbt2bY71Zp8D2ft05513Ovw8PHnypPH19TV9+vRxWC4pKcn4+Pg4tBfVZyUrK8vUqVPHxMTEOJyjZ86cMTVq1DD33HNPgdcFlEbcmgjcoNLS0iRJXl5ehVquefPmqlu3rv11ZmamFi1apA4dOqhmzZr29sqVK+uRRx7RypUr7dvy9fXVli1btHPnzlzX7e7uLhcXFy1btqxAt/hcrGvXrjpy5IjDkPFz5sxRVlaWunbt6rCNbOfOndOxY8d0++23S1Khb4FZsGCBKleurM6dO9vbypUrZ/9r/cUu3u7p06d17NgxNW3aVMYYbdiwoVDblQr3vmfr27evwxW/u+66S5mZmdq3b1+e2ynIMVm8eLFOnjypYcOG5XjmKXt769at05EjR9SvXz+HPu3atVNoaGiuzyI+9dRTDq/j4uLk4+Oje+65R8eOHbNPERER8vT01NKlS/Pcj+x9yZbfsff19dWaNWt06NChfNdXWJd+dnKr68SJE0pNTdVdd91V4PPxSo5rtl69ejlcRbnrrrskSX/++aekf47b33//rT59+jg819StWzeVL1/+suvPvsr57bffKisrK9c+cXFxCgsLU2hoqMNxbdmypSRd9rjmt+2//vpLa9euvaLlMzMzFRsbq5MnT+qbb76xP2N1tedhXqKjox2u8N1yyy3y9va2H4usrCzNmzdP999/vxo1apRj+Uuv5vfp08fh+bfFixcrJSVFsbGxDnWXKVNGTZo0cai7qD4rv/32m3bu3KlHHnlEf//9t32bp0+fVqtWrbR8+XL7eVFcnzvgWkYQA25Q3t7ekv65L78watSo4fD66NGjOnPmjG6++eYcfcPCwpSVlaUDBw5Ikl555RWlpKTopptuUv369TV06FD9/vvv9v6urq4aO3asfvjhBwUEBKhZs2YaN26ckpKS7H1SU1OVlJRkn44fPy5J9ucPLr4da9asWWrYsKFuuukme9vx48c1ePBgBQQEyN3dXf7+/vZ9Sk1NLdR7sW/fPtWuXTvHL0C5vRf79+9Xz549VaFCBXl6esrf399+G2VhtysV7n3PVq1aNYfX2b9I5xd6C3JMdu/eLUn22z9zkx0Kcqs3NDQ0R2hwdnbOcdvbzp07lZqaqkqVKsnf399hOnXqlI4cOZLn9qWCH/tx48Zp8+bNCg4OVmRkpEaNGmX/ZfhqXPrZyTZ//nzdfvvtcnNzU4UKFeTv76/333+/wOfFlRzXgi6bfVwuvV3P2dm5QCNDdu3aVXfccYcef/xxBQQE6OGHH9bs2bMdQtnOnTu1ZcuWHMc0+3N7ueOalxdeeEGenp6KjIxUnTp11L9//xzPJeXnpZde0pIlSzRz5kyHgHS152FeLj0W0j/HI/tYHD16VGlpafl+zi526fmW/Qewli1b5qh70aJFDnUX1Wcle5s9evTIsc2PP/5Y6enp9vUV1+cOuJbxjBhwg/L29lZQUJB9EIuCuvgvpYXVrFkz7d69W99++60WLVqkjz/+WBMmTNC0adP0+OOPS5Kefvpp3X///Zo3b55+/PFHvfzyyxozZoyWLFmiW2+9VYMHD9ann35qX2fz5s21bNkyubq6qkOHDvrmm280depUJScna9WqVXrjjTccaujSpYtWr16toUOHqmHDhvL09FRWVpbatGmT51/sr1ZmZqbuueceHT9+XC+88IJCQ0Pl4eGhgwcPqmfPnsW23UvlNTqcucy3mFzumBQHV1dXOTk5/q0wKytLlSpV0hdffJHrMhc/j5Wbgh77Ll266K677tI333yjRYsWafz48Ro7dqy+/vprtW3b9or3KbfPzooVK/TAAw+oWbNmmjp1qipXrqyyZcvqk08+yTHQSV6u9Lhe7bIF4e7uruXLl2vp0qX6/vvvtXDhQs2aNUstW7bUokWLVKZMGWVlZal+/fp65513cl1HcHDwFW07LCxMO3bs0Pz587Vw4ULNnTtXU6dO1YgRIzR69Oh8l503b57Gjh2rV199VW3atHGYd7XnYV6K+lhcer5ln+MzZsxQYGBgjv4XX/Esqs9Kdt/x48erYcOGudbp6elZoHUBpRFBDLiB3Xffffrwww+VkJCgqKioK1qHv7+/ypUrpx07duSYt337djk5OTn8IlWhQgX16tVLvXr10qlTp9SsWTONGjXKHsSkfx7AHzJkiIYMGaKdO3eqYcOGevvtt/X555/r+eef16OPPmrve/HtUV27dtWnn36q+Ph4bdu2TcYYh9sST5w4ofj4eI0ePVojRoywt+d1q+TlVK9eXZs3b5YxxuGq2KXvxaZNm/THH3/o008/Vffu3e3tF48Wma2gg4UU9n2/Wvkdk+yrBZs3b85x5SRb9erVJf3z3mTfcpZtx44d9vmXq+Gnn37SHXfcUeg/CBT22FeuXFn9+vVTv379dOTIEd122216/fXX8/2F8Eq+h2ru3Llyc3PTjz/+6DA8/SeffFLodRWH7OOya9cu3X333fb2CxcuaO/evbrlllsuuw4nJye1atVKrVq10jvvvKM33nhDL774opYuXWq/HW/jxo1q1arVZd/Dwr7HHh4e6tq1q7p27aqMjAx17NhRr7/+uoYPH57nVwf88ccf6tGjhzp06JBjdEfp6s7Dq+Hv7y9vb+9C//EsW/bntFKlSoqOjs6zX1F+VrK36e3tne82C7IuoDTi1kTgBvb888/Lw8NDjz/+uJKTk3PM3717tyZNmpTvOsqUKaPWrVvr22+/dRjOPDk5WTNnztSdd95pvw3y0qGPPT09Vbt2bfsQ1WfOnNG5c+cc+tSqVUteXl72PnXr1lV0dLR9ioiIsPeNjo5WhQoVNGvWLM2aNUuRkZEOt+dk/8X50r8wT5w4Md99zMu9996rQ4cOOQw1fubMmRxfgJ3bdo0xub632c+hpKSk5LvtwrzvV6Mgx6R169by8vLSmDFjcvTN3udGjRqpUqVKmjZtmsOQ5D/88IO2bdtWoNEBu3TposzMTL366qs55l24cCHf96ygxz4zMzPHLYGVKlVSUFBQrl9LcLGCHrtL67LZbMrMzLS37d271z5SYElr1KiR/Pz89NFHH+nChQv29i+++KJAtz5m3zp8sewrI9nvZ5cuXXTw4EF99NFHOfqePXvWPuKf9M97XND399KfNy4uLqpbt66MMTp//nyuy5w6dUoPPvigqlSpYh8m/1JXcx5eDScnJ3Xo0EHfffed1q1bl2P+5a6cxcTEyNvbW2+88Uau+5/9lQdF+VmJiIhQrVq19NZbb+nUqVN5bvNqPnfA9YwrYsANrFatWpo5c6a6du2qsLAwde/eXfXq1VNGRoZWr16tuLi4y35PjyS99tpr9u8K6tevn5ydnfXBBx8oPT1d48aNs/erW7euWrRooYiICFWoUEHr1q2zD1cs/fOX6FatWqlLly6qW7eunJ2d9c033yg5OVkPP/zwZesoW7asOnbsqK+++kqnT5/WW2+95TDf29vb/ozT+fPnVaVKFS1atEh79uwp3Bv3//Xp00eTJ09W9+7dlZiYqMqVK2vGjBn2odazhYaGqlatWnruued08OBBeXt7a+7cubn+IpsdLAcNGqSYmBiVKVMmz30v6Pt+NQpyTLy9vTVhwgQ9/vjjaty4sR555BGVL19eGzdu1JkzZ/Tpp5+qbNmyGjt2rHr16qXmzZsrNjbWPnx9SEiInnnmmcvW0rx5cz3xxBMaM2aMfvvtN7Vu3Vply5bVzp07FRcXp0mTJjkMnHKxgh77kydPqmrVqurcubMaNGggT09P/fTTT1q7dq3efvvtfOvLPnYvvviiHn74YZUtW1b3339/vl+k265dO73zzjtq06aNHnnkER05ckRTpkxR7dq1HZ6fLCkuLi4aNWqUBg4cqJYtW6pLly7au3evpk+frlq1al32CtUrr7yi5cuXq127dqpevbqOHDmiqVOnqmrVqrrzzjslSY899phmz56tJ598UkuXLtUdd9yhzMxMbd++XbNnz9aPP/5oH5wiIiJCP/30k9555x0FBQWpRo0aOb6mIVvr1q0VGBioO+64QwEBAdq2bZsmT56sdu3a5TlI0ejRo7V161a99NJL+vbbbx3m1apVS1FRUVd1Hl6tN954Q4sWLVLz5s3tQ/0fPnxYcXFxWrlypcNXQFzK29tb77//vh577DHddtttevjhh+Xv76/9+/fr+++/1x133KHJkycX6WfFyclJH3/8sdq2bavw8HD16tVLVapU0cGDB7V06VJ5e3vru+++u6rPHXBdK4GRGgFcY/744w/Tp08fExISYlxcXIyXl5e54447zHvvvWfOnTtn7ycpz+GF169fb2JiYoynp6cpV66cufvuu83q1asd+rz22msmMjLS+Pr6Gnd3dxMaGmpef/11k5GRYYwx5tixY6Z///4mNDTUeHh4GB8fH9OkSROH4eEvZ/HixUaSsdls5sCBAznm//XXX+bBBx80vr6+xsfHxzz00EPm0KFDOYaGL8jw9cYYs2/fPvPAAw+YcuXKmYoVK5rBgwfbh7G+ePj6rVu3mujoaOPp6WkqVqxo+vTpYx+e+pNPPrH3u3Dhghk4cKDx9/c3NpvNYZjrS2s0pmDve/a+XDrkdfYQ5hfXeanCHJP//e9/pmnTpsbd3d14e3ubyMhI8+WXXzr0mTVrlrn11luNq6urqVChgunWrZv566+/HPr06NHDeHh45FnThx9+aCIiIoy7u7vx8vIy9evXN88//7w5dOhQnssYU7Bjn56eboYOHWoaNGhgvLy8jIeHh2nQoIGZOnVqvuvO9uqrr5oqVaoYJycnh/Mnv8/Of/7zH1OnTh3j6upqQkNDzSeffJLrEOd5DV9fkOOa1/D1cXFxDsvu2bMnxzlpjDHvvvuuqV69unF1dTWRkZFm1apVJiIiwrRp0ybf9yM+Pt60b9/eBAUFGRcXFxMUFGRiY2PNH3/84dAvIyPDjB071oSHhxtXV1dTvnx5ExERYUaPHm1SU1Pt/bZv326aNWtm3N3djaR8h7L/4IMPTLNmzYyfn59xdXU1tWrVMkOHDnVY36Wf8+yvTshtunRbV3oe5jV8fW7nx6XH3Jh/fuZ0797d+Pv7G1dXV1OzZk3Tv39/+9cQ5HVeZFu6dKmJiYkxPj4+xs3NzdSqVcv07NnTrFu3zt6nqD8rGzZsMB07drQfi+rVq5suXbqY+Pj4Qq8LKE1sxhTRE7kAAOCGkJWVJX9/f3Xs2DHXWwoBAJfHM2IAACBP586dy/G80Geffabjx4+rRYsWJVMUAJQCXBEDAAB5WrZsmZ555hk99NBD8vPz0/r16/Wf//xHYWFhSkxMdPhCaABAwTFYBwAAyFNISIiCg4P17rvv6vjx46pQoYK6d++uN998kxAGAFeBK2IAAAAAYDGeEQMAAAAAixHEAAAAAMBiPCNWBLKysnTo0CF5eXld9sstAQAAAJRexhidPHlSQUFBcnLK+7oXQawIHDp0SMHBwSVdBgAAAIBrxIEDB1S1atU85xPEioCXl5ekf95sb2/vEq4GAAAAQElJS0tTcHCwPSPkhSBWBLJvR/T29iaIAQAAALjsI0sM1gEAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWu+6C2JQpUxQSEiI3Nzc1adJEv/76a7794+LiFBoaKjc3N9WvX18LFizIs++TTz4pm82miRMnFnHVAAAAAPB/rqsgNmvWLD377LMaOXKk1q9frwYNGigmJkZHjhzJtf/q1asVGxur3r17a8OGDerQoYM6dOigzZs35+j7zTff6JdfflFQUFBx7wYAAACAG9x1FcTeeecd9enTR7169VLdunU1bdo0lStXTv/9739z7T9p0iS1adNGQ4cOVVhYmF599VXddtttmjx5skO/gwcPauDAgfriiy9UtmxZK3YFAAAAwA3sugliGRkZSkxMVHR0tL3NyclJ0dHRSkhIyHWZhIQEh/6SFBMT49A/KytLjz32mIYOHarw8PAC1ZKenq60tDSHCQAAAAAK6roJYseOHVNmZqYCAgIc2gMCApSUlJTrMklJSZftP3bsWDk7O2vQoEEFrmXMmDHy8fGxT8HBwYXYEwAAAAA3uusmiBWHxMRETZo0SdOnT5fNZivwcsOHD1dqaqp9OnDgQDFWCQAAAKC0uW6CWMWKFVWmTBklJyc7tCcnJyswMDDXZQIDA/Ptv2LFCh05ckTVqlWTs7OznJ2dtW/fPg0ZMkQhISF51uLq6ipvb2+HCQAAAAAK6roJYi4uLoqIiFB8fLy9LSsrS/Hx8YqKisp1maioKIf+krR48WJ7/8cee0y///67fvvtN/sUFBSkoUOH6scffyy+nQEAAABwQ3Mu6QIK49lnn1WPHj3UqFEjRUZGauLEiTp9+rR69eolSerevbuqVKmiMWPGSJIGDx6s5s2b6+2331a7du301Vdfad26dfrwww8lSX5+fvLz83PYRtmyZRUYGKibb77Z2p0DAAAAcMO4roJY165ddfToUY0YMUJJSUlq2LChFi5caB+QY//+/XJy+r+LfE2bNtXMmTP10ksv6d///rfq1KmjefPmqV69eiW1CwAAAAAgmzHGlHQR17u0tDT5+PgoNTWV58UAAACAG1hBs8F184wYAAAAAJQWBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACx23QWxKVOmKCQkRG5ubmrSpIl+/fXXfPvHxcUpNDRUbm5uql+/vhYsWGCfd/78eb3wwguqX7++PDw8FBQUpO7du+vQoUPFvRsAAAAAbmDXVRCbNWuWnn32WY0cOVLr169XgwYNFBMToyNHjuTaf/Xq1YqNjVXv3r21YcMGdejQQR06dNDmzZslSWfOnNH69ev18ssva/369fr666+1Y8cOPfDAA1buFgAAAIAbjM0YY0q6iIJq0qSJGjdurMmTJ0uSsrKyFBwcrIEDB2rYsGE5+nft2lWnT5/W/Pnz7W233367GjZsqGnTpuW6jbVr1yoyMlL79u1TtWrVClRXWlqafHx8lJqaKm9v7yvYMwAAAAClQUGzwXVzRSwjI0OJiYmKjo62tzk5OSk6OloJCQm5LpOQkODQX5JiYmLy7C9Jqampstls8vX1zbNPenq60tLSHCYAAAAAKKjrJogdO3ZMmZmZCggIcGgPCAhQUlJSrsskJSUVqv+5c+f0wgsvKDY2Nt/0OmbMGPn4+Nin4ODgQu4NAAAAgBvZdRPEitv58+fVpUsXGWP0/vvv59t3+PDhSk1NtU8HDhywqEoAAAAApYFzSRdQUBUrVlSZMmWUnJzs0J6cnKzAwMBclwkMDCxQ/+wQtm/fPi1ZsuSyz3m5urrK1dX1CvYCAAAAAK6jK2IuLi6KiIhQfHy8vS0rK0vx8fGKiorKdZmoqCiH/pK0ePFih/7ZIWznzp366aef5OfnVzw7AAAAAAD/33VzRUySnn32WfXo0UONGjVSZGSkJk6cqNOnT6tXr16SpO7du6tKlSoaM2aMJGnw4MFq3ry53n77bbVr105fffWV1q1bpw8//FDSPyGsc+fOWr9+vebPn6/MzEz782MVKlSQi4tLyewoAAAAgFLtugpiXbt21dGjRzVixAglJSWpYcOGWrhwoX1Ajv3798vJ6f8u8jVt2lQzZ87USy+9pH//+9+qU6eO5s2bp3r16kmSDh48qP/973+SpIYNGzpsa+nSpWrRooUl+wUAAADgxnJdfY/YtYrvEQMAAAAglcLvEQMAAACA0oIgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWIwgBgAAAAAWI4gBAAAAgMUIYgAAAABgMYIYAAAAAFiMIAYAAAAAFiOIAQAAAIDFrjqIpaWlad68edq2bVtR1AMAAAAApV6hg1iXLl00efJkSdLZs2fVqFEjdenSRbfccovmzp1b5AUCAAAAQGlT6CC2fPly3XXXXZKkb775RsYYpaSk6N1339Vrr71W5AUCAAAAQGlT6CCWmpqqChUqSJIWLlyoTp06qVy5cmrXrp127txZ5AUCAAAAQGlT6CAWHByshIQEnT59WgsXLlTr1q0lSSdOnJCbm1uRFwgAAAAApY1zYRd4+umn1a1bN3l6eqpatWpq0aKFpH9uWaxfv35R1wcAAAAApU6hg1i/fv0UGRmpAwcO6J577pGT0z8X1WrWrMkzYgAAAABQADZjjLmSBTMyMrRnzx7VqlVLzs6FznOlSlpamnx8fJSamipvb++SLgcAAABACSloNij0M2JnzpxR7969Va5cOYWHh2v//v2SpIEDB+rNN9+88ooBAAAA4AZR6CA2fPhwbdy4UcuWLXMYnCM6OlqzZs0q0uIAAAAAoDQq9D2F8+bN06xZs3T77bfLZrPZ28PDw7V79+4iLQ4AAAAASqNCXxE7evSoKlWqlKP99OnTDsEMAAAAAJC7QgexRo0a6fvvv7e/zg5fH3/8saKiooquMgAAAAAopQp9a+Ibb7yhtm3bauvWrbpw4YImTZqkrVu3avXq1fr555+Lo0YAAAAAKFUKfUXszjvv1MaNG3XhwgXVr19fixYtUqVKlZSQkKCIiIjiqBEAAAAASpVCXRE7f/68nnjiCb388sv66KOPiqsmAAAAACjVCnVFrGzZspo7d25x1QIAAAAAN4RC35rYoUMHzZs3rxhKAQAAAIAbQ6EH66hTp45eeeUVrVq1ShEREfLw8HCYP2jQoCIrDgAAAABKI5sxxhRmgRo1auS9MptNf/7551UXdb1JS0uTj4+PUlNT5e3tXdLlAAAAACghBc0Ghb4itmfPnqsqDAAAAABudIV+RuxixhgV8oIaAAAAANzwriiIffbZZ6pfv77c3d3l7u6uW265RTNmzCjq2gAAAACgVCr0rYnvvPOOXn75ZQ0YMEB33HGHJGnlypV68skndezYMT3zzDNFXiQAAAAAlCZXNFjH6NGj1b17d4f2Tz/9VKNGjbohnyFjsA4AAAAAUsGzQaFvTTx8+LCaNm2ao71p06Y6fPhwYVcHAAAAADecQgex2rVra/bs2TnaZ82apTp16hRJUQAAAABQmhX6GbHRo0era9euWr58uf0ZsVWrVik+Pj7XgAYAAAAAcFToK2KdOnXSmjVrVLFiRc2bN0/z5s1TxYoV9euvv+rBBx8sjhodTJkyRSEhIXJzc1OTJk3066+/5ts/Li5OoaGhcnNzU/369bVgwQKH+cYYjRgxQpUrV5a7u7uio6O1c+fO4twFAAAAADe4Kxq+PiIiQp9//rkSExOVmJiozz//XLfeemtR15bDrFmz9Oyzz2rkyJFav369GjRooJiYGB05ciTX/qtXr1ZsbKx69+6tDRs2qEOHDurQoYM2b95s7zNu3Di9++67mjZtmtasWSMPDw/FxMTo3Llzxb4/AAAAAG5MhR41ccGCBSpTpoxiYmIc2n/88UdlZWWpbdu2RVrgxZo0aaLGjRtr8uTJkqSsrCwFBwdr4MCBGjZsWI7+Xbt21enTpzV//nx72+23366GDRtq2rRpMsYoKChIQ4YM0XPPPSdJSk1NVUBAgKZPn66HH364QHUxaiIAAAAAqRhHTRw2bJgyMzNztBtjcg1DRSUjI0OJiYmKjo62tzk5OSk6OloJCQm5LpOQkODQX5JiYmLs/ffs2aOkpCSHPj4+PmrSpEme65Sk9PR0paWlOUwAAAAAUFCFDmI7d+5U3bp1c7SHhoZq165dRVJUbo4dO6bMzEwFBAQ4tAcEBCgpKSnXZZKSkvLtn/1vYdYpSWPGjJGPj499Cg4OLvT+AAAAALhxFTqI+fj46M8//8zRvmvXLnl4eBRJUde64cOHKzU11T4dOHCgpEsCAAAAcB0pdBBr3769nn76ae3evdvetmvXLg0ZMkQPPPBAkRZ3sYoVK6pMmTJKTk52aE9OTlZgYGCuywQGBubbP/vfwqxTklxdXeXt7e0wAQAAAEBBFTqIjRs3Th4eHgoNDVWNGjVUo0YNhYWFyc/PT2+99VZx1ChJcnFxUUREhOLj4+1tWVlZio+PV1RUVK7LREVFOfSXpMWLF9v716hRQ4GBgQ590tLStGbNmjzXCQAAAABXq9Bf6Ozj46PVq1dr8eLF2rhxo9zd3XXLLbeoWbNmxVGfg2effVY9evRQo0aNFBkZqYkTJ+r06dPq1auXJKl79+6qUqWKxowZI0kaPHiwmjdvrrffflvt2rXTV199pXXr1unDDz+UJNlsNj399NN67bXXVKdOHdWoUUMvv/yygoKC1KFDh2LfHwAAAAA3pkIHMemfANO6dWu1bt1akpSSklKUNeWpa9euOnr0qEaMGKGkpCQ1bNhQCxcutA+2sX//fjk5/d9FvqZNm2rmzJl66aWX9O9//1t16tTRvHnzVK9ePXuf559/XqdPn1bfvn2VkpKiO++8UwsXLpSbm5sl+wQAAADgxlPo7xEbO3asQkJC1LVrV0lSly5dNHfuXAUGBmrBggVq0KBBsRR6LeN7xAAAAABIxfg9YtOmTbMP17548WItXrxYP/zwg9q2bauhQ4deecUAAAAAcIMo9K2JSUlJ9iA2f/58denSRa1bt1ZISIiaNGlS5AUCAAAAQGlT6Cti5cuXt39v1sKFCxUdHS1JMsYoMzOzaKsDAAAAgFKo0FfEOnbsqEceeUR16tTR33//rbZt20qSNmzYoNq1axd5gQAAAABQ2hQ6iE2YMEEhISE6cOCAxo0bJ09PT0nS4cOH1a9fvyIvEAAAAABKm0KPmoicGDURAAAAgFSMoyYCAAAAAK4OQQwAAAAALEYQAwAAAACLFTiIMTQ9AAAAABSNAgexKlWqaNiwYfrjjz+Ksx4AAAAAKPUKHMT69++vOXPmKCwsTHfddZemT5+uM2fOFGdtAAAAAFAqFTiIvfzyy9q1a5fi4+NVs2ZNDRgwQJUrV1afPn20Zs2a4qwRAAAAAEqVQg/W0aJFC3366adKSkrS22+/rW3btikqKkrh4eF65513iqNGAAAAAChViuQLnb///nt1795dKSkpN+SgHnyhMwAAAADJgi90PnPmjKZPn67mzZvrgQcekJ+fn15//fUrXR0AAAAA3DCcC7vA6tWr9d///ldxcXG6cOGCOnfurFdffVXNmjUrjvoAAAAAoNQpcBAbN26cPvnkE/3xxx9q1KiRxo8fr9jYWHl5eRVnfQAAAABQ6hQ4iI0fP16PPvqo4uLiVK9eveKsCQAAAABKtQIHsUOHDqls2bLFWQsAAAAA3BAKPFjHihUrVLduXaWlpeWYl5qaqvDwcK1YsaJIiwMAAACA0qjAQWzixInq06dPrkMw+vj46IknnuB7xAAAAACgAAocxDZu3Kg2bdrkOb9169ZKTEwskqIAAAAAoDQrcBBLTk7O9xkxZ2dnHT16tEiKAgAAAIDSrMBBrEqVKtq8eXOe83///XdVrly5SIoCAAAAgNKswEHs3nvv1csvv6xz587lmHf27FmNHDlS9913X5EWBwAAAAClkc0YYwrSMTk5WbfddpvKlCmjAQMG6Oabb5Ykbd++XVOmTFFmZqbWr1+vgICAYi34WpSWliYfHx+lpqbmOpgJAAAAgBtDQbNBgb9HLCAgQKtXr9ZTTz2l4cOHKzu/2Ww2xcTEaMqUKTdkCAMAAACAwipwEJOk6tWra8GCBTpx4oR27dolY4zq1Kmj8uXLF1d9AAAAAFDqFCqIZStfvrwaN25c1LUAAAAAwA2hwIN1AAAAAACKBkEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsNh1E8SOHz+ubt26ydvbW76+vurdu7dOnTqV7zLnzp1T//795efnJ09PT3Xq1EnJycn2+Rs3blRsbKyCg4Pl7u6usLAwTZo0qbh3BQAAAMAN7roJYt26ddOWLVu0ePFizZ8/X8uXL1ffvn3zXeaZZ57Rd999p7i4OP388886dOiQOnbsaJ+fmJioSpUq6fPPP9eWLVv04osvavjw4Zo8eXJx7w4AAACAG5jNGGNKuojL2bZtm+rWrau1a9eqUaNGkqSFCxfq3nvv1V9//aWgoKAcy6Smpsrf318zZ85U586dJUnbt29XWFiYEhISdPvtt+e6rf79+2vbtm1asmRJgetLS0uTj4+PUlNT5e3tfQV7CAAAAKA0KGg2uC6uiCUkJMjX19cewiQpOjpaTk5OWrNmTa7LJCYm6vz584qOjra3hYaGqlq1akpISMhzW6mpqapQoUK+9aSnpystLc1hAgAAAICCui6CWFJSkipVquTQ5uzsrAoVKigpKSnPZVxcXOTr6+vQHhAQkOcyq1ev1qxZsy57y+OYMWPk4+Njn4KDgwu+MwAAAABueCUaxIYNGyabzZbvtH37dktq2bx5s9q3b6+RI0eqdevW+fYdPny4UlNT7dOBAwcsqREAAABA6eBckhsfMmSIevbsmW+fmjVrKjAwUEeOHHFov3Dhgo4fP67AwMBclwsMDFRGRoZSUlIcroolJyfnWGbr1q1q1aqV+vbtq5deeumydbu6usrV1fWy/QAAAAAgNyUaxPz9/eXv73/ZflFRUUpJSVFiYqIiIiIkSUuWLFFWVpaaNGmS6zIREREqW7as4uPj1alTJ0nSjh07tH//fkVFRdn7bdmyRS1btlSPHj30+uuvF8FeAQAAAED+rotREyWpbdu2Sk5O1rRp03T+/Hn16tVLjRo10syZMyVJBw8eVKtWrfTZZ58pMjJSkvTUU09pwYIFmj59ury9vTVw4EBJ/zwLJv1zO2LLli0VExOj8ePH27dVpkyZAgXEbIyaCAAAAEAqeDYo0StihfHFF19owIABatWqlZycnNSpUye9++679vnnz5/Xjh07dObMGXvbhAkT7H3T09MVExOjqVOn2ufPmTNHR48e1eeff67PP//c3l69enXt3bvXkv0CAAAAcOO5bq6IXcu4IgYAAABAKmXfIwYAAAAApQlBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAi103Qez48ePq1q2bvL295evrq969e+vUqVP5LnPu3Dn1799ffn5+8vT0VKdOnZScnJxr37///ltVq1aVzWZTSkpKMewBAAAAAPzjugli3bp105YtW7R48WLNnz9fy5cvV9++ffNd5plnntF3332nuLg4/fzzzzp06JA6duyYa9/evXvrlltuKY7SAQAAAMCBzRhjSrqIy9m2bZvq1q2rtWvXqlGjRpKkhQsX6t5779Vff/2loKCgHMukpqbK399fM2fOVOfOnSVJ27dvV1hYmBISEnT77bfb+77//vuaNWuWRowYoVatWunEiRPy9fUtcH1paWny8fFRamqqvL29r25nAQAAAFy3CpoNrosrYgkJCfL19bWHMEmKjo6Wk5OT1qxZk+syiYmJOn/+vKKjo+1toaGhqlatmhISEuxtW7du1SuvvKLPPvtMTk4FezvS09OVlpbmMAEAAABAQV0XQSwpKUmVKlVyaHN2dlaFChWUlJSU5zIuLi45rmwFBATYl0lPT1dsbKzGjx+vatWqFbieMWPGyMfHxz4FBwcXbocAAAAA3NBKNIgNGzZMNpst32n79u3Ftv3hw4crLCxMjz76aKGXS01NtU8HDhwopgoBAAAAlEbOJbnxIUOGqGfPnvn2qVmzpgIDA3XkyBGH9gsXLuj48eMKDAzMdbnAwEBlZGQoJSXF4apYcnKyfZklS5Zo06ZNmjNnjiQp+3G5ihUr6sUXX9To0aNzXberq6tcXV0LsosAAAAAkEOJBjF/f3/5+/tftl9UVJRSUlKUmJioiIgISf+EqKysLDVp0iTXZSIiIlS2bFnFx8erU6dOkqQdO3Zo//79ioqKkiTNnTtXZ8+etS+zdu1a/etf/9KKFStUq1atq909AAAAAMhViQaxggoLC1ObNm3Up08fTZs2TefPn9eAAQP08MMP20dMPHjwoFq1aqXPPvtMkZGR8vHxUe/evfXss8+qQoUK8vb21sCBAxUVFWUfMfHSsHXs2DH79gozaiIAAAAAFMZ1EcQk6YsvvtCAAQPUqlUrOTk5qVOnTnr33Xft88+fP68dO3bozJkz9rYJEybY+6anpysmJkZTp04tifIBAAAAwO66+B6xax3fIwYAAABAKmXfIwYAAAAApQlBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAAAAACxGEAMAAAAAixHEAAAAAMBiBDEAAAAAsBhBDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwmHNJF1AaGGMkSWlpaSVcCQAAAICSlJ0JsjNCXghiReDkyZOSpODg4BKuBAAAAMC14OTJk/Lx8clzvs1cLqrhsrKysnTo0CF5eXnJZrOVdDnIRVpamoKDg3XgwAF5e3uXdDm4DnDOoLA4Z1BYnDMoLM6Z64MxRidPnlRQUJCcnPJ+EowrYkXAyclJVatWLekyUADe3t784EKhcM6gsDhnUFicMygszplrX35XwrIxWAcAAAAAWIwgBgAAAAAWI4jhhuDq6qqRI0fK1dW1pEvBdYJzBoXFOYPC4pxBYXHOlC4M1gEAAAAAFuOKGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghhKjePHj6tbt27y9vaWr6+vevfurVOnTuW7zLlz59S/f3/5+fnJ09NTnTp1UnJycq59//77b1WtWlU2m00pKSnFsAewUnGcLxs3blRsbKyCg4Pl7u6usLAwTZo0qbh3BcVoypQpCgkJkZubm5o0aaJff/013/5xcXEKDQ2Vm5ub6tevrwULFjjMN8ZoxIgRqly5stzd3RUdHa2dO3cW5y7AQkV5vpw/f14vvPCC6tevLw8PDwUFBal79+46dOhQce8GLFTUP2Mu9uSTT8pms2nixIlFXDWKjAFKiTZt2pgGDRqYX375xaxYscLUrl3bxMbG5rvMk08+aYKDg018fLxZt26duf32203Tpk1z7du+fXvTtm1bI8mcOHGiGPYAViqO8+U///mPGTRokFm2bJnZvXu3mTFjhnF3dzfvvfdece8OisFXX31lXFxczH//+1+zZcsW06dPH+Pr62uSk5Nz7b9q1SpTpkwZM27cOLN161bz0ksvmbJly5pNmzbZ+7z55pvGx8fHzJs3z2zcuNE88MADpkaNGubs2bNW7RaKSVGfLykpKSY6OtrMmjXLbN++3SQkJJjIyEgTERFh5W6hGBXHz5hsX3/9tWnQoIEJCgoyEyZMKOY9wZUiiKFU2Lp1q5Fk1q5da2/74YcfjM1mMwcPHsx1mZSUFFO2bFkTFxdnb9u2bZuRZBISEhz6Tp061TRv3tzEx8cTxEqB4j5fLtavXz9z9913F13xsExkZKTp37+//XVmZqYJCgoyY8aMybV/ly5dTLt27RzamjRpYp544gljjDFZWVkmMDDQjB8/3j4/JSXFuLq6mi+//LIY9gBWKurzJTe//vqrkWT27dtXNEWjRBXXOfPXX3+ZKlWqmM2bN5vq1asTxK5h3JqIUiEhIUG+vr5q1KiRvS06OlpOTk5as2ZNrsskJibq/Pnzio6OtreFhoaqWrVqSkhIsLdt3bpVr7zyij777DM5OfGRKQ2K83y5VGpqqipUqFB0xcMSGRkZSkxMdDjeTk5Oio6OzvN4JyQkOPSXpJiYGHv/PXv2KCkpyaGPj4+PmjRpku85hGtfcZwvuUlNTZXNZpOvr2+R1I2SU1znTFZWlh577DENHTpU4eHhxVM8igy/VaJUSEpKUqVKlRzanJ2dVaFCBSUlJeW5jIuLS47/oQUEBNiXSU9PV2xsrMaPH69q1aoVS+2wXnGdL5davXq1Zs2apb59+xZJ3bDOsWPHlJmZqYCAAIf2/I53UlJSvv2z/y3MOnF9KI7z5VLnzp3TCy+8oNjYWHl7exdN4SgxxXXOjB07Vs7Ozho0aFDRF40iRxDDNW3YsGGy2Wz5Ttu3by+27Q8fPlxhYWF69NFHi20bKDolfb5cbPPmzWrfvr1Gjhyp1q1bW7JNAKXT+fPn1aVLFxlj9P7775d0ObhGJSYmatKkSZo+fbpsNltJl4MCcC7pAoD8DBkyRD179sy3T82aNRUYGKgjR444tF+4cEHHjx9XYGBgrssFBgYqIyNDKSkpDlc5kpOT7cssWbJEmzZt0pw5cyT9M+KZJFWsWFEvvviiRo8efYV7huJQ0udLtq1bt6pVq1bq27evXnrppSvaF5SsihUrqkyZMjlGUc3teGcLDAzMt3/2v8nJyapcubJDn4YNGxZh9bBacZwv2bJD2L59+7RkyRKuhpUSxXHOrFixQkeOHHG4gyczM1NDhgzRxIkTtXfv3qLdCVw1rojhmubv76/Q0NB8JxcXF0VFRSklJUWJiYn2ZZcsWaKsrCw1adIk13VHRESobNmyio+Pt7ft2LFD+/fvV1RUlCRp7ty52rhxo3777Tf99ttv+vjjjyX988Ouf//+xbjnuBIlfb5I0pYtW3T33XerR48eev3114tvZ1GsXFxcFBER4XC8s7KyFB8f73C8LxYVFeXQX5IWL15s71+jRg0FBgY69ElLS9OaNWvyXCeuD8Vxvkj/F8J27typn376SX5+fsWzA7BccZwzjz32mH7//Xf77yy//fabgoKCNHToUP3444/FtzO4ciU9WghQVNq0aWNuvfVWs2bNGrNy5UpTp04dh+HI//rrL3PzzTebNWvW2NuefPJJU61aNbNkyRKzbt06ExUVZaKiovLcxtKlSxk1sZQojvNl06ZNxt/f3zz66KPm8OHD9unIkSOW7huKxldffWVcXV3N9OnTzdatW03fvn2Nr6+vSUpKMsYY89hjj5lhw4bZ+69atco4Ozubt956y2zbts2MHDky1+HrfX19zbfffmt+//130759e4avLyWK+nzJyMgwDzzwgKlatar57bffHH6mpKenl8g+omgVx8+YSzFq4rWNIIZS4++//zaxsbHG09PTeHt7m169epmTJ0/a5+/Zs8dIMkuXLrW3nT171vTr18+UL1/elCtXzjz44IPm8OHDeW6DIFZ6FMf5MnLkSCMpx1S9enUL9wxF6b333jPVqlUzLi4uJjIy0vzyyy/2ec2bNzc9evRw6D979mxz0003GRcXFxMeHm6+//57h/lZWVnm5ZdfNgEBAcbV1dW0atXK7Nixw4pdgQWK8nzJ/hmU23TxzyVc34r6Z8ylCGLXNpsx//+hFwAAAACAJXhGDAAAAAAsRhADAAAAAIsRxAAAAADAYgQxAAAAALAYQQwAAAAALEYQAwAAAACLEcQAAAAAwGIEMQAAAACwGEEMAHDdCwkJ0cSJEwvcf9myZbLZbEpJSSm2mq4VLVq00NNPP13SZQAALmEzxpiSLgIAcGOw2Wz5zh85cqRGjRpV6PUePXpUHh4eKleuXIH6Z2Rk6Pjx4woICLhsTSWpRYsWatiwYaFC5qWOHz+usmXLysvLq+gKAwBcNeeSLgAAcOM4fPiw/b9nzZqlESNGaMeOHfY2T09P+38bY5SZmSln58v/r8rf379Qdbi4uCgwMLBQy1yvKlSoUNIlAABywa2JAADLBAYG2icfHx/ZbDb76+3bt8vLy0s//PCDIiIi5OrqqpUrV2r37t1q3769AgIC5OnpqcaNG+unn35yWO+ltybabDZ9/PHHevDBB1WuXDnVqVNH//vf/+zzL701cfr06fL19dWPP/6osLAweXp6qk2bNg7B8cKFCxo0aJB8fX3l5+enF154QT169FCHDh3y3N99+/bp/vvvV/ny5eXh4aHw8HAtWLDAPn/z5s1q27atPD09FRAQoMcee0zHjh2TJPXs2VM///yzJk2aJJvNJpvNpr179+a6nalTp6pOnTpyc3NTQECAOnfubJ938a2J2ft96dSzZ097/2+//Va33Xab3NzcVLNmTY0ePVoXLlzIcx8BAFeGIAYAuKYMGzZMb775prZt26ZbbrlFp06d0r333qv4+Hht2LBBbdq00f3336/9+/fnu57Ro0erS5cu+v3333XvvfeqW7duOn78eJ79z5w5o7feekszZszQ8uXLtX//fj333HP2+WPHjtUXX3yhTz75RKtWrVJaWprmzZuXbw39+/dXenq6li9frk2bNmns2LH2q34pKSlq2bKlbr31Vq1bt04LFy5UcnKyunTpIkmaNGmSoqKi1KdPHx0+fFiHDx9WcHBwjm2sW7dOgwYN0iuvvKIdO3Zo4cKFatasWa71NG3a1L6uw4cPa8mSJXJzc7P3X7Fihbp3767Bgwdr69at+uCDDzR9+nS9/vrr+e4nAOAKGAAASsAnn3xifHx87K+XLl1qJJl58+Zddtnw8HDz3nvv2V9Xr17dTJgwwf5aknnppZfsr0+dOmUkmR9++MFhWydOnLDXIsns2rXLvsyUKVNMQECA/XVAQIAZP368/fWFCxdMtWrVTPv27fOss379+mbUqFG5znv11VdN69atHdoOHDhgJJkdO3YYY4xp3ry5GTx4cJ7rN8aYuXPnGm9vb5OWlpbr/LzWcezYMVOzZk3Tr18/e1urVq3MG2+84dBvxowZpnLlyvnWAAAoPJ4RAwBcUxo1auTw+tSpUxo1apS+//57HT58WBcuXNDZs2cve0Xslltusf+3h4eHvL29deTIkTz7lytXTrVq1bK/rly5sr1/amqqkpOTFRkZaZ9fpkwZRUREKCsrK891Dho0SE899ZQWLVqk6OhoderUyV7Xxo0btXTpUofn4rLt3r1bN910U777l+2ee+5R9erVVbNmTbVp00Zt2rSx35KZl/Pnz6tTp06qXr26Jk2aZG/fuHGjVq1a5XAFLDMzU+fOndOZM2cKPBgKAODyuDURAHBN8fDwcHj93HPP6ZtvvtEbb7yhFStW6LffflP9+vWVkZGR73rKli3r8Npms+UbmnLrb65yYOHHH39cf/75px577DFt2rRJjRo10nvvvSfpn4B5//3367fffnOYdu7cmeethbnx8vLS+vXr9eWXX6py5coaMWKEGjRokO/Q/E899ZQOHDiguLg4h8FQTp06pdGjRzvUs2nTJu3cuVNubm5X/D4AAHIiiAEArmmrVq1Sz5499eCDD6p+/foKDAzMc9CK4uLj46OAgACtXbvW3paZman169dfdtng4GA9+eST+vrrrzVkyBB99NFHkqTbbrtNW7ZsUUhIiGrXru0wZYdRFxcXZWZmXnYbzs7Oio6O1rhx4/T7779r7969WrJkSa5933nnHc2ePVvffvut/Pz8HObddttt2rFjR456ateuLScnfmUAgKLErYkAgGtanTp19PXXX+v++++XzWbTyy+/nO+VreIycOBAjRkzRrVr11ZoaKjee+89nThxIt/vIXv66afVtm1b3XTTTTpx4oSWLl2qsLAwSf8M5PHRRx8pNjZWzz//vCpUqKBdu3bpq6++0scff6wyZcooJCREa9as0d69e+Xp6akKFSrkCETz58/Xn3/+qWbNmql8+fJasGCBsrKydPPNN+eo56efftLzzz+vKVOmqGLFikpKSpIkubu7y8fHRyNGjNB9992natWqqXPnznJyctLGjRu1efNmvfbaa0X4bgIA+PMWAOCa9s4776h8+fJq2rSp7r//fsXExOi2226zvI4XXnhBsbGx6t69u6KiouTp6amYmJh8b9nLzMxU//79FRYWpjZt2uimm27S1KlTJUlBQUFatWqVMjMz1bp1a9WvX19PP/20fH197WHrueeeU5kyZVS3bl35+/vn+lycr6+vvv76a7Vs2VJhYWGaNm2avvzyS4WHh+fou3LlSmVmZurJJ59U5cqV7dPgwYMlSTExMZo/f74WLVqkxo0b6/bbb9eECRNUvXr1ongLAQAXsZmrvQEeAIAbUFZWlsLCwtSlSxe9+uqrJV0OAOA6w62JAAAUwL59+7Ro0SI1b95c6enpmjx5svbs2aNHHnmkpEsDAFyHuDURAIACcHJy0vTp09W4cWPdcccd2rRpk3766Sf7M18AABQGtyYCAAAAgMW4IgYAAAAAFiOIAQAAAIDFCGIAAAAAYDGCGAAAAABYjCAGAAAAABYjiAEAAACAxQhiAAAAAGAxghgAAAAAWOz/ASSXvnGzX7cAAAAAAElFTkSuQmCC\n" + }, + "metadata": {} + } + ], + "source": [ + "plt.subplots(figsize=(10, 5))\n", + "plt.errorbar(train_size, test_scores_mean, yerr=test_scores_std)\n", + "plt.xlabel('Training set size')\n", + "plt.ylabel('CV scores')\n", + "plt.title('Cross-validation score as training set size increases');" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "GThy0cZ5GC6M" + }, + "source": [ + "This shows that you seem to have plenty of data. There's an initial rapid improvement in model scores as one would expect, but it's essentially levelled off by around a sample size of 40-50." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "MgpNV-4iGC6M" + }, + "source": [ + "## 4.13 Save best model object from pipeline" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "3h74xIAyGC6N" + }, + "outputs": [], + "source": [ + "#Code task 28#\n", + "#This may not be \"production grade ML deployment\" practice, but adding some basic\n", + "#information to your saved models can save your bacon in development.\n", + "#Just what version model have you just loaded to reuse? What version of `sklearn`\n", + "#created it? When did you make it?\n", + "#Assign the pandas version number (`pd.__version__`) to the `pandas_version` attribute,\n", + "#the numpy version (`np.__version__`) to the `numpy_version` attribute,\n", + "#the sklearn version (`sklearn_version`) to the `sklearn_version` attribute,\n", + "#and the current datetime (`datetime.datetime.now()`) to the `build_datetime` attribute\n", + "#Let's call this model version '1.0'\n", + "best_model = rf_grid_cv.best_estimator_\n", + "best_model.version = 1.0\n", + "best_model.pandas_version = pd.__version__\n", + "best_model.numpy_version = np.__version__\n", + "best_model.sklearn_version = sklearn_version\n", + "best_model.X_columns = [col for col in X_train.columns]\n", + "best_model.build_datetime = datetime.datetime.now()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "yyMGrAm5GC6N", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "c9e79227-967d-4992-8e0c-3f75ac024af8" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Directory ../models was created.\n", + "Writing file. \"../models/ski_resort_pricing_model.pkl\"\n" + ] + } + ], + "source": [ + "# save the model\n", + "\n", + "modelpath = '../models'\n", + "save_file(best_model, 'ski_resort_pricing_model.pkl', modelpath)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4UUQ_abTGC6N" + }, + "source": [ + "## 4.14 Summary" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fMyhMImVGC6N" + }, + "source": [ + "**Q: 1** Write a summary of the work in this notebook. Capture the fact that you gained a baseline idea of performance by simply taking the average price and how well that did. Then highlight that you built a linear model and the features that found. Comment on the estimate of its performance from cross-validation and whether its performance on the test split was consistent with this estimate. Also highlight that a random forest regressor was tried, what preprocessing steps were found to be best, and again what its estimated performance via cross-validation was and whether its performance on the test set was consistent with that. State which model you have decided to use going forwards and why. This summary should provide a quick overview for someone wanting to know quickly why the given model was chosen for the next part of the business problem to help guide important business decisions." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "bt8Qg_aGGC6N" + }, + "source": [ + "**A: 1** A baseline idea of performance was gained by simply taking the average ticket price, however, that prediction was found to be within 19 dollars of the real ticket price. To get even closer to the real ticket price, a linear regression model was used and that model explains over 80% of the variance on the train set as well as over 70% on the test set. Using this model, on average, you'd expect to estimate a ticket price within approximately 9 dollars of the real price. Testing its performance using the test/split method, as expected, did not hold up consistently. The next model used is the random forest model. This model has an even lower cross-validation estimate, to the real price, by almost 1 dollar. This model also testing consistent estimates with the various performance results. With all of this data, I have chosen to use the random forest model. This decision was made based off the consistency of the models results, and the ability to use this estimate on various areas of data for additional proactive solutions or predictions for conflict resolution." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.9" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": true, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true + }, + "varInspector": { + "cols": { + "lenName": 16, + "lenType": 16, + "lenVar": 40 + }, + "kernels_config": { + "python": { + "delete_cmd_postfix": "", + "delete_cmd_prefix": "del ", + "library": "var_list.py", + "varRefreshCmd": "print(var_dic_list())" + }, + "r": { + "delete_cmd_postfix": ") ", + "delete_cmd_prefix": "rm(", + "library": "var_list.r", + "varRefreshCmd": "cat(var_dic_list()) " + } + }, + "types_to_exclude": [ + "module", + "function", + "builtin_function_or_method", + "instance", + "_Feature" + ], + "window_display": false + }, + "colab": { + "provenance": [], + "include_colab_link": true + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file