Visualisation Walk Through.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7cec1c7c",
   "metadata": {},
   "source": [
    "This iPython notebook makes several graphs as an example. It does not include a full selection of the graphs in the visualisation toolkit"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2a375ac",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e790e6c4",
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "import time\n",
    "from convertbng.util import convert_bng, convert_lonlat\n",
    "import branca\n",
    "import branca.colormap\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from datetime import datetime\n",
    "import math\n",
    "from math import atan2, degrees\n",
    "import random\n",
    "import pickle\n",
    "\n",
    "import re\n",
    "import os\n",
    "import requests\n",
    "import json\n",
    "from urllib.parse import quote\n",
    "\n",
    "import plotly.express as px\n",
    "import plotly.graph_objects as go\n",
    "from plotly.subplots import make_subplots\n",
    "\n",
    "import ipywidgets as widgets\n",
    "from IPython.display import display\n",
    "from geopy.geocoders import Nominatim\n",
    "import folium\n",
    "from ipywidgets import interact, Dropdown\n",
    "from folium.plugins import MarkerCluster\n",
    "from folium import LinearColormap\n",
    "\n",
    "\n",
    "import geopandas as gpd\n",
    "import contextily as ctx\n",
    "from ipywidgets import interact, SelectionSlider\n",
    "\n",
    "import chardet\n",
    "\n",
    "import branca\n",
    "import networkx as nx\n",
    "\n",
    "import matplotlib.cm as cm\n",
    "import matplotlib.colors as colors\n",
    "from matplotlib.lines import Line2D\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from mpl_toolkits.axes_grid1 import make_axes_locatable"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4763fab2",
   "metadata": {},
   "source": [
    "## Load Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "636fc13e",
   "metadata": {},
   "source": [
    "Unpickle NUMBAT data to load it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "b7d74c7a",
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "#First set the directory\n",
    "directoryfilepath = \"/NUMBAT-main/data\"\n",
    "os.chdir(directoryfilepath)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9ea93471",
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "#Load data (deserialize)\n",
    "with open('datadict.pickle', 'rb') as handle:\n",
    "    data_dict = pickle.load(handle)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba1baf4f",
   "metadata": {},
   "source": [
    "## Map Over London to Visualize magnitudes and changes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "accfac22",
   "metadata": {},
   "source": [
    "Visualise a single period of time over a map of London.\n",
    "Pick the dataframe, column of interst, and direction of travel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "97508347",
   "metadata": {
    "code_folding": [
     1,
     54,
     61,
     71
    ]
   },
   "outputs": [],
   "source": [
    "\n",
    "def oneday_with_nodes(dfone, column_name, mode='total'):\n",
    "    assert mode in ['NIE', 'SOW', 'total'], \"Invalid mode. Choose from 'NIE', 'SOW', or 'total'.\"\n",
    "    df1 = dfone.copy()\n",
    "    \n",
    "    #Filter the dataframe based on direction\n",
    "    #If the direction is total, flip stations in North, In, or East directions and add to their compliment\n",
    "    if mode == \"total\":\n",
    "        df1[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df1.apply(\n",
    "            lambda row: pd.Series(\n",
    "                [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "            ),\n",
    "            axis=1\n",
    "        )\n",
    "\n",
    "        \n",
    "        df1 = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()\n",
    "        \n",
    "    elif mode == 'NIE':\n",
    "        df1 = df1[df1['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "\n",
    "    elif mode == 'SOW':\n",
    "        df1 = df1[df1['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "\n",
    "\n",
    "    # Merge both dataframes and fill missing values with 0\n",
    "    merged_df = df1\n",
    "    merged_df.fillna(0, inplace=True)\n",
    "\n",
    "\n",
    "    merged_df = merged_df.groupby([\"From Station\", \"To Station\", \"StartLat\",\"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()\n",
    "    \n",
    "    plot_df = merged_df[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', \"Hex\", column_name]]\n",
    "\n",
    "    \n",
    "    #Create a background map\n",
    "    london_coords = [51.5074, -0.1278]\n",
    "    map_london = folium.Map(location=london_coords, zoom_start=12, tiles='cartodb positron')\n",
    "    \n",
    "    #Set line width multiple\n",
    "    lineFactor = 15/max(abs(plot_df[column_name]))\n",
    "    \n",
    "    for _, row in plot_df.iterrows():\n",
    "        \n",
    "        #weight of lines\n",
    "        color = \"#\" + row.at[\"Hex\"]\n",
    "        \n",
    "    \n",
    "        linewidth = abs(row[column_name]) * lineFactor\n",
    "            \n",
    "        #add line from start lat to stop lat\n",
    "        folium.PolyLine(\n",
    "            locations=[[row['StartLat'], row['StartLong']], [row['StopLat'], row['StopLong']]],\n",
    "            color=color,\n",
    "            weight=linewidth\n",
    "        ).add_to(map_london)\n",
    "\n",
    "        # Add station nodes and popup text\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=5,\n",
    "            color='black',\n",
    "            fill=True,\n",
    "            fill_color='black'\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=10,  # Increase this value to make the clickable area larger\n",
    "            color='transparent',\n",
    "            fill=True,\n",
    "            fill_opacity=0.0\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "    return map_london\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6354ae6b",
   "metadata": {},
   "source": [
    "Look at traffic in all directions during AMPM Peak in Friday 2020 without any background "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4983b478",
   "metadata": {},
   "outputs": [],
   "source": [
    "oneday_with_nodes(data_dict.get(\"data20FRI\"),  \"AMPM Peak\", mode='total')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70c0662f",
   "metadata": {},
   "source": [
    "Compare two different columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "id": "6862051d",
   "metadata": {
    "code_folding": [
     2,
     31,
     34
    ]
   },
   "outputs": [],
   "source": [
    "#Compare two times with colour scheme\n",
    "#2nd minus 1st\n",
    "def two_times_dfs_with_colour(dfone, dftwo, column_one, column_two, mode='total'):\n",
    "    assert mode in ['NIE', 'SOW', 'total'], \"Invalid mode. Choose from 'NIE', 'SOW', or 'total'.\"\n",
    "    df1 = dfone.copy()\n",
    "    df2 = dftwo.copy()\n",
    "    \n",
    "    #Filter the dataframe based on direction\n",
    "    #If the direction is total, flip stations in North, In, or East directions and add to their compliment\n",
    "    if mode == \"total\":\n",
    "        df1[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df1.apply(\n",
    "            lambda row: pd.Series(\n",
    "                [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "            ),\n",
    "            axis=1\n",
    "        )\n",
    "        df2[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df2.apply(\n",
    "            lambda row: pd.Series(\n",
    "                [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "            ),\n",
    "            axis=1\n",
    "        )\n",
    "        \n",
    "        df1 = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_one: \"sum\"}).reset_index()\n",
    "        df2 = df2.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_two: \"sum\"}).reset_index()    \n",
    "    elif mode == 'NIE':\n",
    "        df1 = df1[df1['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "        df2 = df2[df2['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "    elif mode == 'SOW':\n",
    "        df1 = df1[df1['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "        df2 = df2[df2['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "    df1 = df1[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', \"Hex\",column_one]]\n",
    "    df2 = df2[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', \"Hex\",column_two]]\n",
    "\n",
    "    # Merge both dataframes and fill missing values with 0\n",
    "    merged_df = pd.merge(df1, df2, on=['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', \"Hex\"], how='outer')\n",
    "    merged_df.fillna(0, inplace=True)\n",
    "\n",
    "    # Calculate the difference and set missing routes to df2-0 or 0-df1\n",
    "    merged_df['Difference'] = merged_df[column_two] - merged_df[column_one]\n",
    "            \n",
    "    merged_df = merged_df.groupby([\"From Station\", \"To Station\", \"StartLat\",\"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({\"Difference\": \"sum\"}).reset_index()\n",
    "    \n",
    "    \n",
    "    plot_df = merged_df[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', 'Difference', \"Hex\"]]\n",
    "\n",
    "\n",
    "    london_coords = [51.5074, -0.1278]\n",
    "    map_london = folium.Map(location=london_coords, zoom_start=12, tiles='cartodb positron')\n",
    "    \n",
    "    #Set line width multiple\n",
    "    lineFactor = 10/max(abs(plot_df['Difference'])+ .0001) #0 division problem\n",
    "\n",
    "    for _, row in plot_df.iterrows():\n",
    "        \n",
    "        #Set colours of lines\n",
    "        color = \"#\" + row.at[\"Hex\"]\n",
    "        \n",
    "        linewidth = abs(row['Difference']) * lineFactor\n",
    "            \n",
    "        #Add lines from starting coordinates to end coordinates\n",
    "        folium.PolyLine(\n",
    "            locations=[[row['StartLat'], row['StartLong']], [row['StopLat'], row['StopLong']]],\n",
    "            color=color,\n",
    "            dash_array='5, 10' if row['Difference'] < 0 else None,\n",
    "            weight=linewidth\n",
    "        ).add_to(map_london)\n",
    "\n",
    "        # Add station nodes and popup text\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=5,\n",
    "            color='black',\n",
    "            fill=True,\n",
    "            fill_color='black'\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=10,  # Increase this value to make the clickable area larger\n",
    "            color='transparent',\n",
    "            fill=True,\n",
    "            fill_opacity=0.0\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "    return map_london\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f982170",
   "metadata": {},
   "source": [
    "Compare AM vs PM peak traffic on Saturday 2016 looking at South, Out, and West traffic only with the TfL colour scheme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5c3fc3a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "two_times_dfs_with_colour(data_dict.get(\"data16SAT\"), data_dict.get(\"data16SAT\"), \"AM Peak\", \"PM Peak\", mode='SOW')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9198764",
   "metadata": {},
   "source": [
    "### Add Extra Data for More Advanced Maps\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5d318e9",
   "metadata": {},
   "source": [
    "Download Shape Files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "8478fa74",
   "metadata": {},
   "outputs": [],
   "source": [
    "Directory = \"~/NUMBAT-main/data/\"\n",
    "OA_geo = gpd.read_file(Directory + 'LOAC.shp')\n",
    "borough_geo = gpd.read_file(Directory + 'london_boroughs.geojson')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eea9d39e",
   "metadata": {},
   "source": [
    "Add OA area data from GLA for additional context\n",
    "\n",
    "This data comes from the GLA datastore and is called the MyLondon dataset https://data.london.gov.uk/dataset/mylondon"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "id": "0d797fcd",
   "metadata": {
    "code_folding": [
     4
    ]
   },
   "outputs": [],
   "source": [
    "Directory = \"~/NUMBAT-main/data/\"\n",
    "\n",
    "file_path = Directory + \"MyLondon_LOAC_area_description_text_v3.csv\"\n",
    "\n",
    "with open(file_path, 'rb') as f:\n",
    "    result = chardet.detect(f.read())\n",
    "\n",
    "MyLondon_LOAC_area_description_text_v3 = pd.read_csv(file_path, encoding=result['encoding'])\n",
    "\n",
    "file_path = Directory + 'Summary.csv'\n",
    "transSummary = pd.read_csv(file_path)\n",
    "\n",
    "file_path = Directory + 'modelled_OA_rents.csv'\n",
    "modelled_OA_rents = pd.read_csv(file_path)\n",
    "\n",
    "file_path = Directory + 'MyLondon_fare_zone_OA.csv'\n",
    "MyLondon_fare_zone_OA  = pd.read_csv(file_path)\n",
    "\n",
    "\n",
    "file_path = Directory +'MyLondon_postcode_OA.csv'\n",
    "MyLondon_postcode_OA  = pd.read_csv(file_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4f663a51",
   "metadata": {
    "code_folding": []
   },
   "source": [
    "Get OA and Borough Coordinate Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4c78c2e",
   "metadata": {},
   "source": [
    "## Maps with different Background"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aae21384",
   "metadata": {},
   "source": [
    "Add fare zones to the basic comparison map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "id": "cf140cda",
   "metadata": {
    "code_folding": [
     1
    ]
   },
   "outputs": [],
   "source": [
    "def fare_map_colour_one(dfone, column_name, fare_data, geoson, mode='total', show_station_names=True):\n",
    "    # Check if the mode is valid\n",
    "    assert mode in ['NIE', 'SOW', 'total'], \"Invalid mode. Choose from 'NIE', 'SOW', or 'total'.\"\n",
    "\n",
    "    df1 = dfone.copy()\n",
    "    \n",
    "    # Filter the dataframe based on direction\n",
    "    # If the mode is \"total\", flip stations in North, In, or East directions and add to their complement\n",
    "    if mode == \"total\":\n",
    "        df1[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df1.apply(\n",
    "            lambda row: pd.Series(\n",
    "                [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "            ),\n",
    "            axis=1\n",
    "        )\n",
    "\n",
    "        df1 = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()\n",
    "        \n",
    "    elif mode == 'NIE':\n",
    "        df1 = df1[df1['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "    elif mode == 'SOW':\n",
    "        df1 = df1[df1['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "\n",
    "    merged_df = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()\n",
    "    \n",
    "    plot_df = merged_df[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', column_name, \"Hex\"]]\n",
    "    \n",
    "    # Create a color scale for Fare Zones\n",
    "    color_scale = LinearColormap(\n",
    "        ['#1a1a1a', '#404040', '#696969', '#a3a3a3', '#cccccc', '#f5f5f5'],\n",
    "        vmin=1,\n",
    "        vmax=6,\n",
    "        caption='Fare Zones'\n",
    "    )\n",
    "        \n",
    "    # Update the style function to use fare_data\n",
    "    def style_function(feature):\n",
    "        OA11CD = feature['properties']['OA11CD']\n",
    "        fare_zone_data = fare_data.loc[fare_data['OA11CD'] == OA11CD, 'Fare_Zone']\n",
    "\n",
    "        if fare_zone_data.empty:\n",
    "            return {\n",
    "                'fillColor': 'none',\n",
    "                'fillOpacity': 0,\n",
    "                'color': 'none',\n",
    "                'weight': 0\n",
    "            }\n",
    "        else:\n",
    "            fare_zone = fare_zone_data.values[0]\n",
    "            return {\n",
    "                'fillColor': color_scale(fare_zone),\n",
    "                'fillOpacity': 0.5,\n",
    "                'color': 'none',\n",
    "                'weight': 0\n",
    "            }\n",
    "\n",
    "    london_coords = [51.5074, -0.1278]\n",
    "    map_london = folium.Map(location=london_coords, zoom_start=12, tiles='cartodb positron', zoom_control='topright')\n",
    "\n",
    "    # Add choropleth layer for Fare Zones\n",
    "    folium.GeoJson(\n",
    "        geoson,\n",
    "        name=\"Fare Zones\",\n",
    "        style_function=style_function,\n",
    "        tooltip=folium.GeoJsonTooltip(fields=['OA11CD'], aliases=['LSOA Code'])\n",
    "    ).add_to(map_london)\n",
    "\n",
    "    # Cap line width\n",
    "    lineFactor = 15 / max(abs(plot_df['Difference']) + .0001)  # Add .0001 to avoid division by zero\n",
    "\n",
    "    for _, row in plot_df.iterrows():\n",
    "        # Use red lines for negative volume and green for positive\n",
    "        color = 'green' if row['Difference'] > 0 else 'red'\n",
    "\n",
    "        # Cap line width\n",
    "        linewidth = abs(row['Difference']) * lineFactor\n",
    "\n",
    "        # Add PolyLine for each row\n",
    "        folium.PolyLine(\n",
    "            locations=[[row['StartLat'], row['StartLong']], [row['StopLat'], row['StopLong']]],\n",
    "            color=color,\n",
    "            dash_array='1, 15' if row['Difference'] < 0 else None,  # Use dashed lines for negative volume\n",
    "            weight=linewidth\n",
    "        ).add_to(map_london)\n",
    "\n",
    "\n",
    "        # Add station nodes and popup text\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=5,\n",
    "            color='black',\n",
    "            fill=True,\n",
    "            fill_color='black'\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=10,\n",
    "            color='transparent',\n",
    "            fill=True,\n",
    "            fill_opacity=0.0\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "    legend_html = '''\n",
    "    <div style=\"position: fixed; bottom: 10px; right: 20px; width: 150px; background-color: white; z-index:9999; font-size:14px; padding: 10px; border:2px solid grey;\">\n",
    "    <strong>Fare Zones</strong><br>\n",
    "    '''\n",
    "\n",
    "    fare_zones = fare_data[['Fare_Zone']].drop_duplicates().sort_values('Fare_Zone')\n",
    "\n",
    "    for _, row in fare_zones.iterrows():\n",
    "        legend_html += f'<span style=\"color:{color_scale(row[\"Fare_Zone\"])}\">Zone {row[\"Fare_Zone\"]}</span><br>'\n",
    "\n",
    "    legend_html += '</div>'\n",
    "\n",
    "    # Add the legend to the map\n",
    "    map_london.get_root().html.add_child(folium.Element(legend_html))\n",
    "\n",
    "    return map_london\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf4932aa",
   "metadata": {},
   "source": [
    "Look at Normalised Peak traffic Tuesday-Thursday in 2022 over fare zone"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c2ae626",
   "metadata": {},
   "outputs": [],
   "source": [
    "map_london = fare_map_colour_one(data_dict.get(\"data22TWT\"), \"Normalized AMPM Peak\", MyLondon_fare_zone_OA, OA_geo, mode='NIE')\n",
    "map_london"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73a6b249",
   "metadata": {},
   "source": [
    "Look at information at the OA level"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45f2ac3a",
   "metadata": {},
   "source": [
    "Select OACol, a column to use as background. These graphs are highly maliable. They can display a single NUMBAT dataframe and single time of day or compare two of either. You can also select the colour scheme for OA Col."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 314,
   "id": "d94cf10f",
   "metadata": {
    "code_folding": [
     1,
     5,
     17,
     25,
     27,
     120
    ]
   },
   "outputs": [],
   "source": [
    "def OA_transSummary_multtype_tflcolours(dfone, column_name, OA_data, OACol, geoson, colours = ['lightblue','green'], mode='total', dftwo = False, col2 = False):\n",
    "    assert mode in ['NIE', 'SOW', 'total'], \"Invalid mode. Choose from 'NIE', 'SOW', or 'total'.\"\n",
    "\n",
    "    \n",
    "    if OACol == \"Rent_per_m\":\n",
    "        OA_data[OACol] = OA_data[OACol]/max(OA_data[OACol])\n",
    "    \n",
    "    trans_data = OA_data[['OA', OACol]].set_index('OA')\n",
    "\n",
    "    \n",
    "    cmap = branca.colormap.LinearColormap(colors=colours, index=[0,1],vmin=0,vmax=1)\n",
    "\n",
    "    # Define the color map\n",
    "    #cmap = cm.get_cmap('hot')  # Choose the desired color map\n",
    "\n",
    "    # Function to map rent values to colors\n",
    "    def map_to_color(value):\n",
    "        return colors.rgb2hex(cmap(value))\n",
    "\n",
    "    #Strings true, dfs error\n",
    "    df1 = dfone.copy()\n",
    "    \n",
    "    #See how many dfs there are\n",
    "    twodfs = True\n",
    "    try:\n",
    "        df2 = dftwo.copy()    \n",
    "    except:\n",
    "        twodfs = False\n",
    "        \n",
    "    #See how many cols there are\n",
    "    twocols = bool(col2)\n",
    "        \n",
    "    #Filter the dataframe based on direction\n",
    "    #If the direction is total, flip stations in North, In, or East directions and add to their compliment\n",
    "    \n",
    "    if mode == \"total\":\n",
    "        df1[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df1.apply(\n",
    "            lambda row: pd.Series(\n",
    "                [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "            ),\n",
    "            axis=1\n",
    "        )\n",
    "        if twocols:\n",
    "            df1 = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\", col2: \"sum\"}).reset_index()\n",
    "\n",
    "        else:\n",
    "            df1 = df1.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()\n",
    "\n",
    "        if twodfs:\n",
    "            df2[['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']] = df2.apply(\n",
    "                lambda row: pd.Series(\n",
    "                    [row['To Station'], row['From Station'], row['StopLat'], row['StartLat'], row['StopLong'], row['StartLong']]\n",
    "                    if row['Dir'] in ['EB', 'IB', 'IR', 'NB']\n",
    "                    else [row['From Station'], row['To Station'], row['StartLat'], row['StopLat'], row['StartLong'], row['StopLong']],\n",
    "                    index=['From Station', 'To Station', 'StartLat', 'StopLat', 'StartLong', 'StopLong']\n",
    "                ),\n",
    "                axis=1\n",
    "            )\n",
    "\n",
    "            df2 = df2.groupby([\"From Station\", \"To Station\", \"StartLat\", \"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({column_name: \"sum\"}).reset_index()   \n",
    "\n",
    "    elif mode == 'NIE':\n",
    "        df1 = df1[df1['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "        if twodfs:\n",
    "            df2 = df2[df2['Dir'].isin(['EB', 'IB', 'IR', 'NB'])]\n",
    "    elif mode == 'SOW':\n",
    "        df1 = df1[df1['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "        if twodfs:\n",
    "            df2 = df2[df2['Dir'].isin(['OB', 'OR', 'SB', 'WB'])]\n",
    "\n",
    "    # Merge both dataframes and fill missing values with 0\n",
    "    if twodfs:\n",
    "        merged_df = pd.merge(df1, df2, on=['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', \"Hex\"], how='outer', suffixes=('_df1', '_df2'))\n",
    "            \n",
    "            # Calculate the difference and set missing routes to df2-0 or 0-df1\n",
    "        merged_df['Difference'] = merged_df[column_name + '_df2'] - merged_df[column_name + '_df1']\n",
    "        merged_df = merged_df.groupby([\"From Station\", \"To Station\", \"StartLat\",\"StartLong\", \"StopLat\", \"StopLong\", \"Hex\"]).agg({\"Difference\": \"sum\"}).reset_index()\n",
    "\n",
    "        \n",
    "    elif twocols:\n",
    "        merged_df = df1\n",
    "        merged_df['Difference'] = merged_df[col2] - merged_df[column_name]\n",
    "    \n",
    "    else:\n",
    "        merged_df = df1\n",
    "        merged_df['Difference'] =  merged_df[column_name]\n",
    "        \n",
    "    merged_df.fillna(0, inplace=True)\n",
    "\n",
    "       \n",
    "    plot_df = merged_df[['From Station', 'To Station', 'StartLat', 'StartLong', 'StopLat', 'StopLong', 'Difference', \"Hex\"]]\n",
    "\n",
    "\n",
    "    london_coords = [51.5074, -0.1278]\n",
    "    map_london = folium.Map(location=london_coords, zoom_start=12, tiles='cartodb positron', zoom_control='topright')\n",
    "\n",
    "\n",
    "    \n",
    "    # Define the style function\n",
    "    def style_function(feature):\n",
    "        OA11CD = feature['properties']['OA11CD']\n",
    "        trans = trans_data.loc[trans_data.index == OA11CD, OACol]\n",
    "        \n",
    "\n",
    "        \n",
    "\n",
    "        trans_value = trans.values[0]\n",
    "        color = map_to_color(trans_value)  # Map the rent value to a color\n",
    "\n",
    "        return {\n",
    "            'fillColor': color,\n",
    "            'fillOpacity': np.interp(trans_value, [0, 1], [0.1, .9]),\n",
    "            'color': None,\n",
    "            'weight': 1\n",
    "        }\n",
    "        \n",
    "    folium.GeoJson(\n",
    "        geoson,\n",
    "        name=\"Fare Zones\",\n",
    "        style_function=style_function\n",
    "        ).add_to(map_london)\n",
    "\n",
    "    \n",
    "\n",
    "    # Add PolyLines and CircleMarkers for each row in plot_df\n",
    "    tot = sum(abs(plot_df['Difference']))\n",
    "    \n",
    "    lineFactor = 10/max(plot_df['Difference'] + .0001) #0 division problem\n",
    "    for _, row in plot_df.iterrows():\n",
    "        \n",
    "        \n",
    "        color = \"#\" + row.at[\"Hex\"]\n",
    "        \n",
    "        linewidth = abs(row['Difference']) * lineFactor\n",
    "        \n",
    "\n",
    "\n",
    "        folium.PolyLine(\n",
    "            locations=[[row['StartLat'], row['StartLong']], [row['StopLat'], row['StopLong']]],\n",
    "            color=color,\n",
    "            dash_array='1, 15' if row['Difference'] < 0 else None,\n",
    "            weight=linewidth\n",
    "        ).add_to(map_london)\n",
    "\n",
    "        # Add station nodes and popup text\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=5,\n",
    "            color='black',\n",
    "            fill=True,\n",
    "            fill_color='black'\n",
    "        ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "\n",
    "        # Add a larger transparent circle to make it easier to click on the station\n",
    "        folium.CircleMarker(\n",
    "            location=[row['StartLat'], row['StartLong']],\n",
    "            radius=10,        color='transparent',\n",
    "        fill=True,\n",
    "        fill_opacity=0.0\n",
    "    ).add_child(folium.Popup(row['From Station'])).add_to(map_london)\n",
    "        \n",
    "\n",
    "\n",
    "\n",
    "    map_london.add_child(cmap)\n",
    "\n",
    "\n",
    "    return map_london\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b10819f4",
   "metadata": {},
   "source": [
    "Use green space as a background using gray for areas that lack green space and green for OAs with lots of green. Compare AM Peak traffic Monday-Thursday in 2019 with 2020."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "34be71a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "dfone= data_dict.get(\"data19MTT\")\n",
    "dftwo= data_dict.get(\"data20MTT\")\n",
    "column_name = \"AM Peak\"\n",
    "\n",
    "OA_transSummary_multtype_tflcolours(dfone, column_name, transSummary, \"green_spac\", OA_geo, [\"gray\", \"green\"], mode='total', dftwo = dftwo, col2 = False)\n"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}