LSTM_CustomerRevenue.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eb96d3e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the needed packages\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib as mpl\n",
    "from datetime import datetime\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler, MinMaxScaler, RobustScaler, MaxAbsScaler\n",
    "from scipy.spatial import distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f7563fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# read new dataset from AAG\n",
    "#df = pd.read_parquet('cleaned_data.parquet')\n",
    "\n",
    "# Only keep rows in 2023?\n",
    "#df = df[[\"OrderNumber\", \"OrderDate\", \"CustomerID\", \"NetRevenue\", \"Quantity\", \"Season\", \"City\", \"SalesChannel\"]]\n",
    "#df = df[df['OrderDate'].dt.year == 2023]\n",
    "#df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9fd3dfe1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 33 entries, 362 to 11059\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   CustomerID  33 non-null     object\n",
      " 1   cluster     33 non-null     int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 792.0+ bytes\n",
      "None\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 23500514 entries, 0 to 23500513\n",
      "Data columns (total 3 columns):\n",
      " #   Column      Dtype         \n",
      "---  ------      -----         \n",
      " 0   OrderDate   datetime64[ns]\n",
      " 1   CustomerID  object        \n",
      " 2   NetRevenue  float64       \n",
      "dtypes: datetime64[ns](1), float64(1), object(1)\n",
      "memory usage: 537.9+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "#### either run the customer prediction file first or load the needed datasets created before ####\n",
    "\n",
    "# idea: take cluster assignments from customer clustering\n",
    "# within each cluster, fit model to NetRevenue data (2022,2023)\n",
    "# predict future NetRevenue per CustomerID (2024)\n",
    "\n",
    "\n",
    "# get customer - cluster assignment when running the clustering file \n",
    "# ...\n",
    "# clustered_customers_df = top25_scaled[[\"CustomerID\", \"cluster\"]]\n",
    "# clustered_customers_df.to_csv('clusterAssignments.csv', index = False)\n",
    "\n",
    "# get customer - cluster assignment from pre-saved file\n",
    "clustered_customers_df = pd.read_csv('clusterAssignments.csv', sep = \",\") #this file is created after running the clustering\n",
    "clustered_customers_df.CustomerID = clustered_customers_df.CustomerID.astype(str)\n",
    "\n",
    "c_subsets = {}\n",
    "for num in clustered_customers_df['cluster'].unique():\n",
    "    c_subsets[num] = clustered_customers_df[clustered_customers_df['cluster'] == num]\n",
    "\n",
    "# subset per cluster, use smallest cluster for first LSTM approaches\n",
    "print(c_subsets[2].info())\n",
    "\n",
    "\n",
    "allOrders = pd.read_csv(\"top25OrdersRevenue.csv\")\n",
    "allOrders[\"CustomerID\"] = allOrders[\"CustomerID\"].astype(str)\n",
    "allOrders[\"OrderDate\"] = pd.to_datetime(allOrders['OrderDate'], format='%Y-%m-%d')\n",
    "print(allOrders.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "da5cbf12",
   "metadata": {},
   "outputs": [],
   "source": [
    "#print(c_subsets[1].info())\n",
    "#print(allOrders.info())\n",
    "#allOrders[\"CustomerID\"] = allOrders[\"CustomerID\"].astype(str)\n",
    "#print(allOrders.info())\n",
    "\n",
    "## get values from one cluster \n",
    "orders_subsets = allOrders[allOrders['CustomerID'].isin(c_subsets[2]['CustomerID'])]\n",
    "orders_subsets[\"OrderDate\"] = pd.to_datetime(orders_subsets['OrderDate'], format='%Y-%m-%d')\n",
    "print(orders_subsets.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a25ce409",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 489 entries, 7 to 11183\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   CustomerID  489 non-null    object\n",
      " 1   cluster     489 non-null    int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 11.5+ KB\n",
      "None\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 10672 entries, 0 to 11193\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   CustomerID  10672 non-null  object\n",
      " 1   cluster     10672 non-null  int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 250.1+ KB\n",
      "None\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 33 entries, 362 to 11059\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   CustomerID  33 non-null     object\n",
      " 1   cluster     33 non-null     int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 792.0+ bytes\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "#print(orders_subsets.head(10))\n",
    "#print(orders_subsets_cluster3.head(10))\n",
    "#print(orders_subsets_cluster2.head(10))\n",
    "print(c_subsets[0].info())\n",
    "print(c_subsets[1].info())\n",
    "print(c_subsets[2].info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c3eca0d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "customlevelpred[100:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "id": "baea186c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   OrderDate    NetRevenue\n",
      "0 2021-12-27   3061.350785\n",
      "1 2022-01-03  20784.907068\n",
      "2 2022-01-10  12727.215969\n",
      "3 2022-01-17  14527.650524\n",
      "4 2022-01-24  11662.307592\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 130 entries, 0 to 129\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype         \n",
      "---  ------      --------------  -----         \n",
      " 0   OrderDate   130 non-null    datetime64[ns]\n",
      " 1   NetRevenue  130 non-null    float64       \n",
      "dtypes: datetime64[ns](1), float64(1)\n",
      "memory usage: 3.0 KB\n",
      "None\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the time series data\n",
    "customlevelpred = orders_subsets.sort_values(by=\"OrderDate\")\n",
    "# customlevelpred = customlevelpred.groupby('OrderDate')[['NetRevenue', 'Quantity']].sum().reset_index()\n",
    "customlevelpred = customlevelpred.groupby('OrderDate')['NetRevenue'].sum().reset_index()\n",
    "# aggregate on weekly basis\n",
    "#customlevelpred = customlevelpred.groupby([pd.Grouper(key='OrderDate', freq='W-MON', label='left')])[['NetRevenue', 'Quantity']].sum().reset_index().sort_values('OrderDate')\n",
    "customlevelpred = customlevelpred.groupby([pd.Grouper(key='OrderDate', freq='W-MON', label='left')])['NetRevenue'].sum().reset_index().sort_values('OrderDate')\n",
    "\n",
    "# set date as indexing value for time series analyses such as ARIMA\n",
    "#customlevelpred.set_index('OrderDate', inplace=True)\n",
    "customlevelpred = customlevelpred[:130]\n",
    "print(customlevelpred.head())\n",
    "print(customlevelpred.info())\n",
    "\n",
    "# plot the time series\n",
    "plt.plot(customlevelpred[\"OrderDate\"], customlevelpred[\"NetRevenue\"])\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Revenue')\n",
    "plt.xticks(rotation = 45)\n",
    "plt.title('Time Series Data')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e6350d01",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ADF Statistic: -2.3548882842341907\n",
      "p-value: 0.1548755831197295\n",
      "\t1%: -3.483\n",
      "\t5%: -2.885\n",
      "\t10%: -2.579\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# USE THIS ONE for fitting on different functions\n",
    "# TBD handle stationarity, trend, seasonality\n",
    "\n",
    "# https://www.timescale.com/blog/how-to-work-with-time-series-in-python/\n",
    "    \n",
    "# test for stationarity\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "# Assuming 'data' is the time series data\n",
    "result = adfuller(customlevelpred[\"NetRevenue\"])\n",
    "print('ADF Statistic:', result[0])\n",
    "print('p-value:', result[1])\n",
    "for key, value in result[4].items():\n",
    "    print('\\t%s: %.3f' % (key, value))\n",
    "# p-value <= 0.05: Reject the null hypothesis (H0), the data does not have a unit root and is stationary\n",
    "# TBD cluster 3 data: revenue is not stationary, rescale for better predictions\n",
    "\n",
    "# test for autocorrelation\n",
    "from statsmodels.graphics.tsaplots import plot_acf, plot_pacf\n",
    "# Assuming 'data' is the time series data\n",
    "plot_acf(customlevelpred[\"NetRevenue\"])\n",
    "plot_pacf(customlevelpred[\"NetRevenue\"])\n",
    "plt.show()\n",
    "# ACF are within 95% confidence interval = data doesn’t have any autocorrelation\n",
    "# ACF outside of interval = autocorrelation\n",
    "# TBD cluster 3 data: potential autocorrelation between lags 1 to 5, consider either before or during training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dbfa9a86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# TBD handle trend and seasonality\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "decompose_result = seasonal_decompose(customlevelpred[\"NetRevenue\"], period = 52).plot()\n",
    "#decompose_result.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "0fe865eb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'from statsmodels.tsa.statespace.sarimax import SARIMAX\\n\\ndata1=customlevelpred[\"NetRevenue\"]\\n#data1.head()\\ndata2 = customlevelpred[\"Quantity\"]\\nmodel = SARIMAX(data1, exog=data2,  enforce_stationarity=True,  order=(5, 2, 5))\\ncustomlevelpred[\"SARIMAX\"] = model.fit().fittedvalues\\ncustomlevelpred[[\"NetRevenue\",\"SARIMAX\"]].plot(title=\"SARIMAX\")'"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# customer level traditional prediction approaches\n",
    "\n",
    "#import warnings\n",
    "#warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "# TBD further look into traditional time series analysis methods in addition to ML approaches below\n",
    "# https://medium.com/analytics-vidhya/python-code-on-holt-winters-forecasting-3843808a9873\n",
    "\n",
    "# single exponential smoothing\n",
    "from statsmodels.tsa.holtwinters import SimpleExpSmoothing   \n",
    "customlevelpred[\"HWES1\"] = SimpleExpSmoothing(customlevelpred[\"NetRevenue\"]).fit().fittedvalues\n",
    "customlevelpred[[\"NetRevenue\",\"HWES1\"]].plot(title=\"Holt Winters Single Exponential Smoothing\")\n",
    "\n",
    "\n",
    "# double and triple exponential smoothing\n",
    "from statsmodels.tsa.holtwinters import ExponentialSmoothing\n",
    "customlevelpred[\"HWES3_ADD\"] = ExponentialSmoothing(customlevelpred[\"NetRevenue\"],trend=\"add\").fit().fittedvalues\n",
    "customlevelpred[\"HWES3_MUL\"] = ExponentialSmoothing(customlevelpred[\"NetRevenue\"],trend=\"mul\").fit().fittedvalues\n",
    "customlevelpred[[\"NetRevenue\",\"HWES3_ADD\", \"HWES3_MUL\"]].plot(title=\"Holt Winters Triple Exponential Smoothing: Additive and Multiplicative Seasonality\")\n",
    "\n",
    "\n",
    "# following code based on https://machinelearningmastery.com/time-series-forecasting-methods-in-python-cheat-sheet/\n",
    "# ARIMA\n",
    "from statsmodels.tsa.arima.model import ARIMA\n",
    "# fit model\n",
    "model = ARIMA(customlevelpred[\"NetRevenue\"], order=(1, 1, 1))\n",
    "customlevelpred[\"ARIMA\"] = model_fit = model.fit().fittedvalues\n",
    "customlevelpred[[\"NetRevenue\",\"ARIMA\"]].plot(title=\"ARIMA\")\n",
    " \n",
    "# SARIMAX example\n",
    "# plots: best fitting\n",
    "# TBD look into further optimization and potential predictions using SARIMAX\n",
    "from statsmodels.tsa.statespace.sarimax import SARIMAX\n",
    "\n",
    "data1=customlevelpred[\"NetRevenue\"]\n",
    "#data1.head()\n",
    "data2 = customlevelpred[\"Quantity\"]\n",
    "model = SARIMAX(data1, exog=data2,  enforce_stationarity=True,  order=(5, 2, 5))\n",
    "customlevelpred[\"SARIMAX\"] = model.fit().fittedvalues\n",
    "customlevelpred[[\"NetRevenue\",\"SARIMAX\"]].plot(title=\"SARIMAX\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb1c3763",
   "metadata": {},
   "outputs": [],
   "source": [
    "customlevelpred['OrderDate'] = pd.to_datetime(customlevelpred['OrderDate'])\n",
    "customlevelpred.set_index('OrderDate', inplace=True)\n",
    "#print(customlevelpred)\n",
    "\n",
    "# Fit the ARIMA model\n",
    "model = ARIMA(customlevelpred[\"NetRevenue\"], order=(5, 2, 5), seasonal_order=(0, 0, 0, 52), enforce_stationarity = True)\n",
    "model_fit = model.fit()\n",
    "\n",
    "# Forecast\n",
    "forecast_steps = 26\n",
    "forecast = model_fit.forecast(steps=forecast_steps)\n",
    "\n",
    "# Create a DataFrame to hold the forecasted values\n",
    "last_date = customlevelpred.index[-1]\n",
    "forecast_dates = pd.date_range(start=last_date, periods=forecast_steps, freq='W-MON')[1:]\n",
    "forecast_df = pd.DataFrame(forecast, index=forecast_dates)\n",
    "\n",
    "# Combine the original data with the forecast\n",
    "#customlevelpred = pd.DataFrame(customlevelpred['NetRevenue'])\n",
    "customlevelpred[\"ARIMA\"] = model_fit.fittedvalues\n",
    "combined_df = pd.concat([customlevelpred, forecast_df], axis=0)\n",
    "\n",
    "# Plot the original data and the forecast\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.plot(combined_df.index, combined_df[\"NetRevenue\"], label=\"Original Data\")\n",
    "plt.plot(customlevelpred.index, customlevelpred[\"ARIMA\"], label=\"ARIMA Fitted Values\")\n",
    "plt.plot(combined_df.index, combined_df[\"predicted_mean\"], label=\"Forecast\", linestyle='dashed')\n",
    "plt.title(\"2024 - Weekly ARIMA Forecast of Revenue for Customers in Cluster 3\")\n",
    "plt.xlabel(\"Date (weeks)\")\n",
    "plt.ylabel(\"Revenue\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "id": "332e0c6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#combined_df.to_csv(\"2024predictionsCluster3ARIMA.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "id": "b207b70e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NetRevenue</th>\n",
       "      <th>ARIMA</th>\n",
       "      <th>predicted_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2021-12-27</th>\n",
       "      <td>3061.350785</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-03</th>\n",
       "      <td>20784.907068</td>\n",
       "      <td>4254.533084</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-10</th>\n",
       "      <td>12727.215969</td>\n",
       "      <td>23871.533249</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-17</th>\n",
       "      <td>14527.650524</td>\n",
       "      <td>19415.694853</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-24</th>\n",
       "      <td>11662.307592</td>\n",
       "      <td>20490.352909</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-11-11</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>48123.934376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-11-18</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>50549.430785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-11-25</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>50883.548951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-12-02</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>48589.772737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2024-12-09</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>51388.970831</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>155 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              NetRevenue         ARIMA  predicted_mean\n",
       "2021-12-27   3061.350785      0.000000             NaN\n",
       "2022-01-03  20784.907068   4254.533084             NaN\n",
       "2022-01-10  12727.215969  23871.533249             NaN\n",
       "2022-01-17  14527.650524  19415.694853             NaN\n",
       "2022-01-24  11662.307592  20490.352909             NaN\n",
       "...                  ...           ...             ...\n",
       "2024-11-11           NaN           NaN    48123.934376\n",
       "2024-11-18           NaN           NaN    50549.430785\n",
       "2024-11-25           NaN           NaN    50883.548951\n",
       "2024-12-02           NaN           NaN    48589.772737\n",
       "2024-12-09           NaN           NaN    51388.970831\n",
       "\n",
       "[155 rows x 3 columns]"
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "id": "57ce37b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Squared Error (MSE) on scaled data: 0.018698769969589118\n",
      "Mean Absolute Error (MAE) on scaled data: 0.09702146111931456\n"
     ]
    }
   ],
   "source": [
    "# tbd scale to have comparisons\n",
    "\n",
    "import numpy as np\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "\n",
    "# Extract actual and predicted values\n",
    "# Make sure to exclude the forecasted periods from evaluation\n",
    "actual = combined_df.loc[combined_df['NetRevenue'].notna(), 'NetRevenue']\n",
    "predicted = combined_df.loc[combined_df['NetRevenue'].notna(), 'ARIMA']\n",
    "\n",
    "# Reshape the data to fit the scaler\n",
    "actual_reshaped = actual.values.reshape(-1, 1)\n",
    "predicted_reshaped = predicted.values.reshape(-1, 1)\n",
    "\n",
    "# Fit the MinMaxScaler on the actual values\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "scaler.fit(actual_reshaped)\n",
    "\n",
    "# Transform both actual and predicted values\n",
    "actual_scaled = scaler.transform(actual_reshaped)\n",
    "predicted_scaled = scaler.transform(predicted_reshaped)\n",
    "\n",
    "# Calculate metrics on the scaled values\n",
    "mse = mean_squared_error(actual_scaled, predicted_scaled)\n",
    "mae = mean_absolute_error(actual_scaled, predicted_scaled)\n",
    "\n",
    "\n",
    "print(f'Mean Squared Error (MSE) on scaled data: {mse}')\n",
    "print(f'Mean Absolute Error (MAE) on scaled data: {mae}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "id": "ea573aac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NetRevenue</th>\n",
       "      <th>ARIMA</th>\n",
       "      <th>predicted_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2021-12-27</th>\n",
       "      <td>6550.040576</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-03</th>\n",
       "      <td>21021.429319</td>\n",
       "      <td>8974.726196</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-10</th>\n",
       "      <td>26732.985602</td>\n",
       "      <td>25975.170842</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-17</th>\n",
       "      <td>29902.784031</td>\n",
       "      <td>34069.246363</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-01-24</th>\n",
       "      <td>23850.312827</td>\n",
       "      <td>35970.518525</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-11-20</th>\n",
       "      <td>4076.460733</td>\n",
       "      <td>11743.215440</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-11-27</th>\n",
       "      <td>6571.947644</td>\n",
       "      <td>11474.237754</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-12-04</th>\n",
       "      <td>6193.586387</td>\n",
       "      <td>7101.636234</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-12-11</th>\n",
       "      <td>16215.890052</td>\n",
       "      <td>7252.559173</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2023-12-18</th>\n",
       "      <td>11023.663613</td>\n",
       "      <td>12353.638232</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              NetRevenue         ARIMA  predicted_mean\n",
       "2021-12-27   6550.040576      0.000000             NaN\n",
       "2022-01-03  21021.429319   8974.726196             NaN\n",
       "2022-01-10  26732.985602  25975.170842             NaN\n",
       "2022-01-17  29902.784031  34069.246363             NaN\n",
       "2022-01-24  23850.312827  35970.518525             NaN\n",
       "...                  ...           ...             ...\n",
       "2023-11-20   4076.460733  11743.215440             NaN\n",
       "2023-11-27   6571.947644  11474.237754             NaN\n",
       "2023-12-04   6193.586387   7101.636234             NaN\n",
       "2023-12-11  16215.890052   7252.559173             NaN\n",
       "2023-12-18  11023.663613  12353.638232             NaN\n",
       "\n",
       "[104 rows x 3 columns]"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "actual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dc348655",
   "metadata": {},
   "outputs": [],
   "source": [
    "# not needed anymore\n",
    "\n",
    "# tbd make prediction SARIMAX\n",
    "yhat = model_fit.get_prediction(start = \"2024-01-01\", end = \"2024-12-01\")# tbd exog variable??\n",
    "print(yhat)\n",
    "\n",
    "\n",
    "\n",
    "# tried random forest for time series analysis \n",
    "# but realized that this might not be suitable in our context because it does not consider the time aspect\n",
    "\n",
    "# https://swngui.medium.com/python-tutorial-using-random-forest-for-sales-forecasting-bda61d79318e\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "customlevelpred1 = customlevelpred[[\"OrderDate\", \"NetRevenue\", \"Quantity\"]]\n",
    "train_data, test_data = customlevelpred1.loc[customlevelpred1[\"OrderDate\"] < \"2024-01-01\"], customlevelpred1.loc[customlevelpred1[\"OrderDate\"] > \"2024-01-01\"]\n",
    "\n",
    "train_data.info()\n",
    "test_data.info()\n",
    "\n",
    "# Initialize the model\n",
    "rf = RandomForestRegressor(n_estimators=100, random_state=42)\n",
    "\n",
    "# Fit the model on the training data\n",
    "rf.fit(train_data.drop(['OrderDate','NetRevenue'], axis=1), train_data['NetRevenue'])\n",
    "\n",
    "# Predict the sales for the testing data\n",
    "predictions = rf.predict(test_data.drop(['OrderDate','NetRevenue'], axis=1))\n",
    "\n",
    "# Calculate the root mean squared error (RMSE)\n",
    "rmse = ((predictions - test_data['NetRevenue']) ** 2).mean() ** 0.5\n",
    "print('Root Mean Squared Error: {:.2f}'.format(rmse))\n",
    "\n",
    "# Plot the actual and predicted sales\n",
    "plt.scatter(test_data['NetRevenue'], predictions)\n",
    "plt.xlabel('Actual Revenue')\n",
    "plt.ylabel('Predicted Revenue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "be0af26b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 131 entries, 0 to 130\n",
      "Data columns (total 3 columns):\n",
      " #   Column      Non-Null Count  Dtype         \n",
      "---  ------      --------------  -----         \n",
      " 0   OrderDate   131 non-null    datetime64[ns]\n",
      " 1   NetRevenue  131 non-null    float64       \n",
      " 2   Quantity    131 non-null    float64       \n",
      "dtypes: datetime64[ns](1), float64(2)\n",
      "memory usage: 4.1 KB\n"
     ]
    }
   ],
   "source": [
    "# clean data for lstm if Arima models were run before\n",
    "customlevelpred = customlevelpred.drop([\"HWES1\", \"HWES3_ADD\", \"HWES3_MUL\", \"ARIMA\", \"SARIMAX\"], axis = 1)\n",
    "customlevelpred.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "6fb32003",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   OrderDate    NetRevenue\n",
      "0 2021-12-27   3061.350785\n",
      "1 2022-01-03  20784.907068\n",
      "2 2022-01-10  12727.215969\n",
      "3 2022-01-17  14527.650524\n",
      "4 2022-01-24  11662.307592\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 130 entries, 0 to 129\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype         \n",
      "---  ------      --------------  -----         \n",
      " 0   OrderDate   130 non-null    datetime64[ns]\n",
      " 1   NetRevenue  130 non-null    float64       \n",
      "dtypes: datetime64[ns](1), float64(1)\n",
      "memory usage: 3.0 KB\n",
      "None\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "104 26\n"
     ]
    }
   ],
   "source": [
    "# LSTM for time series forecasting DONE\n",
    "# fine tuning, optimal parameter DONE\n",
    "# grid search DONE\n",
    "# find ideal loss function, layer architecture, accuracy measures for time series in our context DONE\n",
    "# TBD use additional input features for more information, ideas: orderCount, Quantities, different ArticleIDs aggregated on cluster level\n",
    "\n",
    "\n",
    "# https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras/\n",
    "# load packages\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import LSTM\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "# get input data and inspect to now basis\n",
    "#lstmdatainput = customlevelpred.drop(\"Quantity\", axis = 1)\n",
    "lstmdatainput = customlevelpred\n",
    "print(lstmdatainput.head())\n",
    "print(lstmdatainput.info())\n",
    "\n",
    "# plot the time series that is considered as input array\n",
    "plt.plot(lstmdatainput[\"OrderDate\"], lstmdatainput['NetRevenue'])\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Revenue')\n",
    "plt.xticks(rotation = 45)\n",
    "plt.title('Time Series Data')\n",
    "plt.show()\n",
    "\n",
    "\n",
    "# random seed for reproducibility\n",
    "tf.random.set_seed(5)\n",
    "\n",
    "# transform integer into float for better handling in NN\n",
    "dataframe = lstmdatainput[\"NetRevenue\"]\n",
    "dataset = dataframe.values\n",
    "dataset = dataset.astype('float32')\n",
    "\n",
    "#print(dataset)\n",
    "\n",
    "# normalize the dataset\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "dataset = scaler.fit_transform(dataset.reshape(-1, 1))\n",
    "\n",
    "#print(dataset)\n",
    "\n",
    "# create trainign and testing split\n",
    "# TBD adapt to be split on 2022/2023 - 2024 (104 weeks training, 27 (of 2024) weeks testing)\n",
    "train_size = int(len(dataset) * 0.8)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]\n",
    "print(len(train), len(test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "244ba520",
   "metadata": {},
   "outputs": [],
   "source": [
    "# function for creating the data input in LSTM format\n",
    "\n",
    "def create_dataset(dataset, look_back=1):\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(dataset)-look_back-1):\n",
    "        a = dataset[i:(i+look_back), 0]\n",
    "        dataX.append(a)\n",
    "        dataY.append(dataset[i + look_back, 0])\n",
    "    return np.array(dataX), np.array(dataY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "91952caf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# has to be rerun before every model training otherwise input format is messed up\n",
    "\n",
    "# TBD find best lag (cf. autocorrelation displayed in lags 1 to 5)\n",
    "# lag 2 already got better results than lag 1, find optimal one\n",
    "look_back = 10\n",
    "trainX, trainY = create_dataset(train, look_back)\n",
    "testX, testY = create_dataset(test, look_back)\n",
    "\n",
    "# reshape input to be [samples, time steps, features]\n",
    "trainX = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))\n",
    "testX = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b52fe404",
   "metadata": {},
   "source": [
    "### grid search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ba64d13b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# grid search to test combinations of parameters to find best settings\n",
    "# takes a while to run with as many paramters to check, should not be necessary to rerun again\n",
    "\n",
    "# reference https://blog.gopenai.com/how-to-perform-grid-search-hyperparameter-tuning-for-lstm-9bed04932d95\n",
    "\n",
    "from keras.wrappers.scikit_learn import KerasRegressor\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# Create the LSTM for grid search\n",
    "def create_lstm_model(neurons=36, second_layer=12, loss='mean_absolute_error', activation='relu', optimizer='Adam', batch_size=12):\n",
    "    model = Sequential()\n",
    "    model.add(LSTM(units=neurons, activation=activation, input_shape=(1, look_back)))\n",
    "    \n",
    "    #if second_layer > 0: \n",
    "    #    model.add(LSTM(units=second_layer, return_sequences=False))\n",
    "    \n",
    "    model.add(Dense(1))\n",
    "    model.compile(loss=loss, optimizer=optimizer, metrics=['mse'])\n",
    "    \n",
    "    return model\n",
    "\n",
    "# Wrap the custom wrapper function for use with scikit-learn\n",
    "lstm_regressor = KerasRegressor(build_fn=create_lstm_model, epochs=50, batch_size=6, verbose=0)\n",
    "\n",
    "# all parameter combinations to test\n",
    "param_grid = {\n",
    "    'optimizer': ['SGD', 'Adam'],\n",
    "    'loss': ['mean_squared_error', 'mean_absolute_error', 'root_mean_squared_error', 'huber'],\n",
    "    'batch_size': [1, 4, 6], # 12\n",
    "    'epochs': [100, 250],\n",
    "    'neurons': [12, 24],\n",
    "    #'second_layer':[0, 12],  # 0 means no second LSTM layer\n",
    "    'activation': ['relu', 'tanh'] #['relu', 'tanh', 'sigmoid', 'linear']\n",
    "}\n",
    "\n",
    "# run the grid search algorithm on the LSTM\n",
    "grid_search = GridSearchCV(estimator=lstm_regressor, param_grid=param_grid, scoring='neg_root_mean_squared_error', cv=3)\n",
    "grid_search_result = grid_search.fit(trainX, trainY)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "ab93096b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                            params  \\\n",
      "63     {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}   \n",
      "15     {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}   \n",
      "127    {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}   \n",
      "29     {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}   \n",
      "51   {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optim...   \n",
      "31     {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}   \n",
      "7    {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'opti...   \n",
      "3    {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optim...   \n",
      "83   {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optim...   \n",
      "125    {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}   \n",
      "\n",
      "     mean_test_score  mean_fit_time  \n",
      "63         -0.148562      16.402237  \n",
      "15         -0.150693      21.628776  \n",
      "127        -0.150796      60.320535  \n",
      "29         -0.150971      51.060780  \n",
      "51         -0.151164      15.901799  \n",
      "31         -0.151185      51.632126  \n",
      "7          -0.151197      21.144575  \n",
      "3          -0.151399      21.042843  \n",
      "83         -0.151511      12.234911  \n",
      "125        -0.151579      59.309108  \n",
      "\n",
      "Best: -0.148562 using {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.203950 (0.014211) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.173772 (0.014721) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.197009 (0.021667) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.151399 (0.023520) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.194204 (0.016913) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.169728 (0.009063) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.175321 (0.016074) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.151197 (0.035593) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.208320 (0.014324) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.172402 (0.014732) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.207383 (0.017614) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.150693 (0.021167) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.207410 (0.027982) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.153383 (0.019201) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.194847 (0.021461) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.153503 (0.018419) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.161787 (0.034272) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.155130 (0.022288) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.171679 (0.028863) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.155096 (0.021569) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.207756 (0.019217) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.150971 (0.020602) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.209160 (0.016611) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.151185 (0.027487) with: {'activation': 'relu', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.219126 (0.021312) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.185882 (0.011779) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.217973 (0.027874) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.186460 (0.013829) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.214349 (0.023781) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.187274 (0.018075) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.213915 (0.020258) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.189871 (0.010628) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.218703 (0.017833) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.183755 (0.011357) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.217196 (0.020926) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.179866 (0.009855) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.213665 (0.017129) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.182220 (0.013384) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.210691 (0.023607) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.151164 (0.029608) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.200220 (0.003464) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.182128 (0.018335) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.209132 (0.011021) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.160298 (0.023213) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.212682 (0.016459) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.154281 (0.017973) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.213606 (0.018336) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.148562 (0.032526) with: {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.220423 (0.023033) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.194894 (0.013115) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.222035 (0.022183) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.184729 (0.011339) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.213172 (0.021727) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.194203 (0.017835) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.211633 (0.027028) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.195886 (0.013368) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.229649 (0.027642) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.192463 (0.012611) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.220795 (0.024834) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.193794 (0.011231) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.211379 (0.018138) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.172985 (0.013217) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.209016 (0.017202) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.151511 (0.019348) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.209932 (0.024148) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.185586 (0.011707) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.198958 (0.012995) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.164252 (0.026763) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.221683 (0.020516) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.180540 (0.020258) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.216404 (0.018021) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.166980 (0.006927) with: {'activation': 'relu', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.193801 (0.017082) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.182772 (0.016295) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.197116 (0.018080) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.180754 (0.018737) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.176728 (0.019866) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.180947 (0.008521) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.178846 (0.021379) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.180858 (0.011836) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.207398 (0.015765) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.181414 (0.017701) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.204764 (0.017512) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.182589 (0.017859) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.191857 (0.016731) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.156555 (0.015988) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.192664 (0.017646) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.152021 (0.019745) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.168577 (0.031839) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.162339 (0.018958) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.170174 (0.032811) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.163391 (0.018573) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.195667 (0.011901) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.151579 (0.019803) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.203816 (0.021908) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.150796 (0.021606) with: {'activation': 'tanh', 'batch_size': 1, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.225109 (0.027920) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.192538 (0.014058) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.208170 (0.014890) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.194593 (0.015960) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.200860 (0.021139) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.187482 (0.019969) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.212334 (0.013630) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.189730 (0.016508) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.220946 (0.023332) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.191778 (0.016248) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.213040 (0.013163) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.192378 (0.015195) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.209889 (0.025294) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.180322 (0.014115) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.206028 (0.014115) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.174743 (0.016877) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.196507 (0.010934) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.178423 (0.017171) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.192787 (0.015885) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.176600 (0.021760) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.206085 (0.018353) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.184297 (0.024120) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.219453 (0.026665) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.176575 (0.017303) with: {'activation': 'tanh', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.230320 (0.040057) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.196496 (0.012704) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.211999 (0.016209) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.197747 (0.014517) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.212294 (0.027331) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.193390 (0.013313) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.199073 (0.012284) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.192753 (0.014928) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.227646 (0.023937) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.193584 (0.013389) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.221450 (0.020952) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.196145 (0.015148) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 100, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.221296 (0.027921) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.186338 (0.014613) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.216578 (0.023179) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.182693 (0.016033) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.192974 (0.017438) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.185349 (0.013525) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.192654 (0.017670) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.183959 (0.016514) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'mean_absolute_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "nan (nan) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'root_mean_squared_error', 'neurons': 24, 'optimizer': 'Adam'}\n",
      "-0.215819 (0.017419) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'SGD'}\n",
      "-0.183795 (0.016578) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 12, 'optimizer': 'Adam'}\n",
      "-0.208886 (0.012074) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'SGD'}\n",
      "-0.183469 (0.014722) with: {'activation': 'tanh', 'batch_size': 6, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}\n"
     ]
    }
   ],
   "source": [
    "# show the 10 best parameter sets\n",
    "results_df = pd.DataFrame(grid_search_result.cv_results_)\n",
    "top_params = results_df.nlargest(10, 'mean_test_score')[['params', 'mean_test_score', 'mean_fit_time']]\n",
    "pd.options.display.max_colwidth = 110\n",
    "print(top_params)\n",
    "\n",
    "#print(top_params.params)\n",
    "#print(top_params.loc[1,\"params\"])\n",
    "\n",
    "print(\"\")\n",
    "print(\"Best: %f using %s\" % (grid_search_result.best_score_, grid_search_result.best_params_))\n",
    "means = grid_search_result.cv_results_['mean_test_score']\n",
    "stds = grid_search_result.cv_results_['std_test_score']\n",
    "params = grid_search_result.cv_results_['params']\n",
    "for mean, stdev, param in zip(means, stds, params):\n",
    "    print(\"%f (%f) with: %r\" % (mean, stdev, param))\n",
    "    \n",
    "#Best: -0.148562 using {'activation': 'relu', 'batch_size': 4, 'epochs': 250, 'loss': 'huber', 'neurons': 24, 'optimizer': 'Adam'}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fb20add",
   "metadata": {},
   "source": [
    "### train model\n",
    "\n",
    "in the following, different parameter settings are tested and evaluated on their prediction quality\n",
    "\n",
    "basis: grid search proposed best parameter settings + literature search on LSTM parameter fine tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "0e3cbc59",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "93/93 - 3s - loss: 0.0319 - mse: 0.0319 - mae: 0.1211 - mape: 1295595.0000 - 3s/epoch - 29ms/step\n",
      "Epoch 2/100\n",
      "93/93 - 0s - loss: 0.0258 - mse: 0.0258 - mae: 0.1074 - mape: 1729736.2500 - 243ms/epoch - 3ms/step\n",
      "Epoch 3/100\n",
      "93/93 - 0s - loss: 0.0255 - mse: 0.0255 - mae: 0.1120 - mape: 1729363.7500 - 240ms/epoch - 3ms/step\n",
      "Epoch 4/100\n",
      "93/93 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1097 - mape: 1666294.5000 - 264ms/epoch - 3ms/step\n",
      "Epoch 5/100\n",
      "93/93 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1097 - mape: 1739856.1250 - 231ms/epoch - 2ms/step\n",
      "Epoch 6/100\n",
      "93/93 - 0s - loss: 0.0247 - mse: 0.0247 - mae: 0.1077 - mape: 1653399.3750 - 273ms/epoch - 3ms/step\n",
      "Epoch 7/100\n",
      "93/93 - 0s - loss: 0.0247 - mse: 0.0247 - mae: 0.1072 - mape: 1705360.3750 - 247ms/epoch - 3ms/step\n",
      "Epoch 8/100\n",
      "93/93 - 0s - loss: 0.0244 - mse: 0.0244 - mae: 0.1098 - mape: 1829413.3750 - 245ms/epoch - 3ms/step\n",
      "Epoch 9/100\n",
      "93/93 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1068 - mape: 1603098.5000 - 254ms/epoch - 3ms/step\n",
      "Epoch 10/100\n",
      "93/93 - 0s - loss: 0.0240 - mse: 0.0240 - mae: 0.1083 - mape: 1799615.6250 - 254ms/epoch - 3ms/step\n",
      "Epoch 11/100\n",
      "93/93 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1061 - mape: 1664511.8750 - 253ms/epoch - 3ms/step\n",
      "Epoch 12/100\n",
      "93/93 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1065 - mape: 1682499.6250 - 251ms/epoch - 3ms/step\n",
      "Epoch 13/100\n",
      "93/93 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1078 - mape: 1676120.2500 - 282ms/epoch - 3ms/step\n",
      "Epoch 14/100\n",
      "93/93 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1049 - mape: 1627541.6250 - 257ms/epoch - 3ms/step\n",
      "Epoch 15/100\n",
      "93/93 - 0s - loss: 0.0234 - mse: 0.0234 - mae: 0.1037 - mape: 1497382.7500 - 251ms/epoch - 3ms/step\n",
      "Epoch 16/100\n",
      "93/93 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1103 - mape: 1850147.2500 - 265ms/epoch - 3ms/step\n",
      "Epoch 17/100\n",
      "93/93 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1039 - mape: 1527241.6250 - 242ms/epoch - 3ms/step\n",
      "Epoch 18/100\n",
      "93/93 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1047 - mape: 1610060.8750 - 225ms/epoch - 2ms/step\n",
      "Epoch 19/100\n",
      "93/93 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1054 - mape: 1638175.6250 - 233ms/epoch - 3ms/step\n",
      "Epoch 20/100\n",
      "93/93 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1052 - mape: 1594587.0000 - 243ms/epoch - 3ms/step\n",
      "Epoch 21/100\n",
      "93/93 - 0s - loss: 0.0232 - mse: 0.0232 - mae: 0.1057 - mape: 1693210.8750 - 232ms/epoch - 2ms/step\n",
      "Epoch 22/100\n",
      "93/93 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1002 - mape: 1440938.2500 - 295ms/epoch - 3ms/step\n",
      "Epoch 23/100\n",
      "93/93 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1066 - mape: 1669933.3750 - 340ms/epoch - 4ms/step\n",
      "Epoch 24/100\n",
      "93/93 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1021 - mape: 1459167.0000 - 354ms/epoch - 4ms/step\n",
      "Epoch 25/100\n",
      "93/93 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1029 - mape: 1501329.3750 - 314ms/epoch - 3ms/step\n",
      "Epoch 26/100\n",
      "93/93 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1058 - mape: 1641799.8750 - 348ms/epoch - 4ms/step\n",
      "Epoch 27/100\n",
      "93/93 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1043 - mape: 1644362.8750 - 339ms/epoch - 4ms/step\n",
      "Epoch 28/100\n",
      "93/93 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1026 - mape: 1616115.7500 - 326ms/epoch - 4ms/step\n",
      "Epoch 29/100\n",
      "93/93 - 0s - loss: 0.0225 - mse: 0.0225 - mae: 0.1055 - mape: 1494966.3750 - 284ms/epoch - 3ms/step\n",
      "Epoch 30/100\n",
      "93/93 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1043 - mape: 1543865.7500 - 325ms/epoch - 3ms/step\n",
      "Epoch 31/100\n",
      "93/93 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1012 - mape: 1596948.8750 - 283ms/epoch - 3ms/step\n",
      "Epoch 32/100\n",
      "93/93 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1026 - mape: 1496896.7500 - 274ms/epoch - 3ms/step\n",
      "Epoch 33/100\n",
      "93/93 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1022 - mape: 1537669.5000 - 273ms/epoch - 3ms/step\n",
      "Epoch 34/100\n",
      "93/93 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1045 - mape: 1569758.6250 - 282ms/epoch - 3ms/step\n",
      "Epoch 35/100\n",
      "93/93 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1047 - mape: 1464247.8750 - 264ms/epoch - 3ms/step\n",
      "Epoch 36/100\n",
      "93/93 - 0s - loss: 0.0227 - mse: 0.0227 - mae: 0.0999 - mape: 1317942.7500 - 264ms/epoch - 3ms/step\n",
      "Epoch 37/100\n",
      "93/93 - 0s - loss: 0.0222 - mse: 0.0222 - mae: 0.1054 - mape: 1740310.2500 - 262ms/epoch - 3ms/step\n",
      "Epoch 38/100\n",
      "93/93 - 0s - loss: 0.0222 - mse: 0.0222 - mae: 0.1016 - mape: 1465678.2500 - 274ms/epoch - 3ms/step\n",
      "Epoch 39/100\n",
      "93/93 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1029 - mape: 1550594.2500 - 272ms/epoch - 3ms/step\n",
      "Epoch 40/100\n",
      "93/93 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1029 - mape: 1501887.8750 - 273ms/epoch - 3ms/step\n",
      "Epoch 41/100\n",
      "93/93 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1018 - mape: 1444143.8750 - 333ms/epoch - 4ms/step\n",
      "Epoch 42/100\n",
      "93/93 - 0s - loss: 0.0221 - mse: 0.0221 - mae: 0.1013 - mape: 1417644.1250 - 323ms/epoch - 3ms/step\n",
      "Epoch 43/100\n",
      "93/93 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1051 - mape: 1529584.5000 - 324ms/epoch - 3ms/step\n",
      "Epoch 44/100\n",
      "93/93 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1051 - mape: 1475079.8750 - 316ms/epoch - 3ms/step\n",
      "Epoch 45/100\n",
      "93/93 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1018 - mape: 1469961.7500 - 306ms/epoch - 3ms/step\n",
      "Epoch 46/100\n",
      "93/93 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1028 - mape: 1445434.1250 - 294ms/epoch - 3ms/step\n",
      "Epoch 47/100\n",
      "93/93 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1020 - mape: 1374666.7500 - 316ms/epoch - 3ms/step\n",
      "Epoch 48/100\n",
      "93/93 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1024 - mape: 1450293.5000 - 323ms/epoch - 3ms/step\n",
      "Epoch 49/100\n",
      "93/93 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1016 - mape: 1479312.1250 - 353ms/epoch - 4ms/step\n",
      "Epoch 50/100\n",
      "93/93 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1043 - mape: 1433729.8750 - 323ms/epoch - 3ms/step\n",
      "Epoch 51/100\n",
      "93/93 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1007 - mape: 1341107.5000 - 364ms/epoch - 4ms/step\n",
      "Epoch 52/100\n",
      "93/93 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1029 - mape: 1447486.6250 - 357ms/epoch - 4ms/step\n",
      "Epoch 53/100\n",
      "93/93 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1031 - mape: 1464196.8750 - 355ms/epoch - 4ms/step\n",
      "Epoch 54/100\n",
      "93/93 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1022 - mape: 1374397.0000 - 345ms/epoch - 4ms/step\n",
      "Epoch 55/100\n",
      "93/93 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1036 - mape: 1303959.8750 - 316ms/epoch - 3ms/step\n",
      "Epoch 56/100\n",
      "93/93 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1033 - mape: 1387399.1250 - 325ms/epoch - 3ms/step\n",
      "Epoch 57/100\n",
      "93/93 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1030 - mape: 1308802.1250 - 348ms/epoch - 4ms/step\n",
      "Epoch 58/100\n",
      "93/93 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1016 - mape: 1399397.8750 - 306ms/epoch - 3ms/step\n",
      "Epoch 59/100\n",
      "93/93 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1025 - mape: 1348286.8750 - 315ms/epoch - 3ms/step\n",
      "Epoch 60/100\n",
      "93/93 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1020 - mape: 1402819.8750 - 397ms/epoch - 4ms/step\n",
      "Epoch 61/100\n",
      "93/93 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1028 - mape: 1302774.6250 - 368ms/epoch - 4ms/step\n",
      "Epoch 62/100\n",
      "93/93 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1012 - mape: 1334263.8750 - 352ms/epoch - 4ms/step\n",
      "Epoch 63/100\n",
      "93/93 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1034 - mape: 1395521.7500 - 348ms/epoch - 4ms/step\n",
      "Epoch 64/100\n",
      "93/93 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1032 - mape: 1388595.7500 - 386ms/epoch - 4ms/step\n",
      "Epoch 65/100\n",
      "93/93 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1023 - mape: 1395407.6250 - 335ms/epoch - 4ms/step\n",
      "Epoch 66/100\n",
      "93/93 - 0s - loss: 0.0215 - mse: 0.0215 - mae: 0.1021 - mape: 1344827.7500 - 378ms/epoch - 4ms/step\n",
      "Epoch 67/100\n",
      "93/93 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.1002 - mape: 1302888.8750 - 319ms/epoch - 3ms/step\n",
      "Epoch 68/100\n",
      "93/93 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1016 - mape: 1274012.8750 - 313ms/epoch - 3ms/step\n",
      "Epoch 69/100\n",
      "93/93 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1032 - mape: 1309685.8750 - 322ms/epoch - 3ms/step\n",
      "Epoch 70/100\n",
      "93/93 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1013 - mape: 1392547.0000 - 352ms/epoch - 4ms/step\n",
      "Epoch 71/100\n",
      "93/93 - 0s - loss: 0.0209 - mse: 0.0209 - mae: 0.1007 - mape: 1317235.5000 - 343ms/epoch - 4ms/step\n",
      "Epoch 72/100\n",
      "93/93 - 0s - loss: 0.0211 - mse: 0.0211 - mae: 0.1007 - mape: 1228368.5000 - 377ms/epoch - 4ms/step\n",
      "Epoch 73/100\n",
      "93/93 - 0s - loss: 0.0209 - mse: 0.0209 - mae: 0.1015 - mape: 1397014.0000 - 306ms/epoch - 3ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "93/93 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.1034 - mape: 1186700.0000 - 366ms/epoch - 4ms/step\n",
      "Epoch 75/100\n",
      "93/93 - 0s - loss: 0.0209 - mse: 0.0209 - mae: 0.1025 - mape: 1435613.6250 - 303ms/epoch - 3ms/step\n",
      "Epoch 76/100\n",
      "93/93 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1020 - mape: 1387202.5000 - 295ms/epoch - 3ms/step\n",
      "Epoch 77/100\n",
      "93/93 - 0s - loss: 0.0207 - mse: 0.0207 - mae: 0.1018 - mape: 1350546.6250 - 301ms/epoch - 3ms/step\n",
      "Epoch 78/100\n",
      "93/93 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1010 - mape: 1234897.6250 - 349ms/epoch - 4ms/step\n",
      "Epoch 79/100\n",
      "93/93 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1008 - mape: 1339445.1250 - 311ms/epoch - 3ms/step\n",
      "Epoch 80/100\n",
      "93/93 - 0s - loss: 0.0207 - mse: 0.0207 - mae: 0.1016 - mape: 1303958.7500 - 367ms/epoch - 4ms/step\n",
      "Epoch 81/100\n",
      "93/93 - 0s - loss: 0.0205 - mse: 0.0205 - mae: 0.1001 - mape: 1204312.3750 - 320ms/epoch - 3ms/step\n",
      "Epoch 82/100\n",
      "93/93 - 0s - loss: 0.0204 - mse: 0.0204 - mae: 0.0996 - mape: 1217637.2500 - 322ms/epoch - 3ms/step\n",
      "Epoch 83/100\n",
      "93/93 - 0s - loss: 0.0205 - mse: 0.0205 - mae: 0.1029 - mape: 1371012.0000 - 356ms/epoch - 4ms/step\n",
      "Epoch 84/100\n",
      "93/93 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1010 - mape: 1344833.6250 - 377ms/epoch - 4ms/step\n",
      "Epoch 85/100\n",
      "93/93 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.0984 - mape: 1192645.6250 - 324ms/epoch - 3ms/step\n",
      "Epoch 86/100\n",
      "93/93 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.1009 - mape: 1259226.0000 - 335ms/epoch - 4ms/step\n",
      "Epoch 87/100\n",
      "93/93 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1004 - mape: 1358055.5000 - 314ms/epoch - 3ms/step\n",
      "Epoch 88/100\n",
      "93/93 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1029 - mape: 1385472.0000 - 326ms/epoch - 4ms/step\n",
      "Epoch 89/100\n",
      "93/93 - 0s - loss: 0.0200 - mse: 0.0200 - mae: 0.0989 - mape: 1263242.3750 - 365ms/epoch - 4ms/step\n",
      "Epoch 90/100\n",
      "93/93 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1015 - mape: 1214989.7500 - 353ms/epoch - 4ms/step\n",
      "Epoch 91/100\n",
      "93/93 - 0s - loss: 0.0204 - mse: 0.0204 - mae: 0.1014 - mape: 1207653.7500 - 421ms/epoch - 5ms/step\n",
      "Epoch 92/100\n",
      "93/93 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.1016 - mape: 1209846.6250 - 375ms/epoch - 4ms/step\n",
      "Epoch 93/100\n",
      "93/93 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.0983 - mape: 1174674.7500 - 334ms/epoch - 4ms/step\n",
      "Epoch 94/100\n",
      "93/93 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.1002 - mape: 1260428.6250 - 335ms/epoch - 4ms/step\n",
      "Epoch 95/100\n",
      "93/93 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.0992 - mape: 1247091.5000 - 316ms/epoch - 3ms/step\n",
      "Epoch 96/100\n",
      "93/93 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.0982 - mape: 1231546.7500 - 345ms/epoch - 4ms/step\n",
      "Epoch 97/100\n",
      "93/93 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.0987 - mape: 1077279.0000 - 331ms/epoch - 4ms/step\n",
      "Epoch 98/100\n",
      "93/93 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.1004 - mape: 1316884.7500 - 328ms/epoch - 4ms/step\n",
      "Epoch 99/100\n",
      "93/93 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.0989 - mape: 1273194.3750 - 328ms/epoch - 4ms/step\n",
      "Epoch 100/100\n",
      "93/93 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.0995 - mape: 1274844.2500 - 354ms/epoch - 4ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x295490a9c30>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 1 from reference above\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(4, input_shape=(1, look_back)))\n",
    "# TBD add second dense layer with more units? or pooling layer??\n",
    "model.add(Dense(1))\n",
    "# TBD choose best loss function \n",
    "model.compile(loss='mean_squared_error', optimizer='adam', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=1, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "9a78d490",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "93/93 - 2s - loss: 0.0342 - mse: 0.0683 - mae: 0.1870 - mape: 908859.6875 - 2s/epoch - 27ms/step\n",
      "Epoch 2/100\n",
      "93/93 - 0s - loss: 0.0163 - mse: 0.0326 - mae: 0.1233 - mape: 1954395.3750 - 252ms/epoch - 3ms/step\n",
      "Epoch 3/100\n",
      "93/93 - 0s - loss: 0.0149 - mse: 0.0298 - mae: 0.1227 - mape: 2099230.5000 - 243ms/epoch - 3ms/step\n",
      "Epoch 4/100\n",
      "93/93 - 0s - loss: 0.0144 - mse: 0.0287 - mae: 0.1205 - mape: 2053375.8750 - 252ms/epoch - 3ms/step\n",
      "Epoch 5/100\n",
      "93/93 - 0s - loss: 0.0142 - mse: 0.0284 - mae: 0.1189 - mape: 2084592.1250 - 244ms/epoch - 3ms/step\n",
      "Epoch 6/100\n",
      "93/93 - 0s - loss: 0.0140 - mse: 0.0280 - mae: 0.1173 - mape: 1979224.2500 - 244ms/epoch - 3ms/step\n",
      "Epoch 7/100\n",
      "93/93 - 0s - loss: 0.0138 - mse: 0.0277 - mae: 0.1164 - mape: 2019213.6250 - 253ms/epoch - 3ms/step\n",
      "Epoch 8/100\n",
      "93/93 - 0s - loss: 0.0135 - mse: 0.0270 - mae: 0.1154 - mape: 2051539.1250 - 261ms/epoch - 3ms/step\n",
      "Epoch 9/100\n",
      "93/93 - 0s - loss: 0.0133 - mse: 0.0266 - mae: 0.1137 - mape: 1867361.5000 - 243ms/epoch - 3ms/step\n",
      "Epoch 10/100\n",
      "93/93 - 0s - loss: 0.0132 - mse: 0.0263 - mae: 0.1150 - mape: 2069297.5000 - 251ms/epoch - 3ms/step\n",
      "Epoch 11/100\n",
      "93/93 - 0s - loss: 0.0130 - mse: 0.0260 - mae: 0.1115 - mape: 1906765.3750 - 243ms/epoch - 3ms/step\n",
      "Epoch 12/100\n",
      "93/93 - 0s - loss: 0.0129 - mse: 0.0258 - mae: 0.1113 - mape: 1900472.2500 - 242ms/epoch - 3ms/step\n",
      "Epoch 13/100\n",
      "93/93 - 0s - loss: 0.0127 - mse: 0.0254 - mae: 0.1132 - mape: 1925369.2500 - 243ms/epoch - 3ms/step\n",
      "Epoch 14/100\n",
      "93/93 - 0s - loss: 0.0126 - mse: 0.0251 - mae: 0.1091 - mape: 1848481.0000 - 231ms/epoch - 2ms/step\n",
      "Epoch 15/100\n",
      "93/93 - 0s - loss: 0.0125 - mse: 0.0249 - mae: 0.1078 - mape: 1685972.1250 - 242ms/epoch - 3ms/step\n",
      "Epoch 16/100\n",
      "93/93 - 0s - loss: 0.0125 - mse: 0.0250 - mae: 0.1150 - mape: 2022296.7500 - 292ms/epoch - 3ms/step\n",
      "Epoch 17/100\n",
      "93/93 - 0s - loss: 0.0125 - mse: 0.0249 - mae: 0.1080 - mape: 1714732.7500 - 305ms/epoch - 3ms/step\n",
      "Epoch 18/100\n",
      "93/93 - 0s - loss: 0.0122 - mse: 0.0243 - mae: 0.1080 - mape: 1771685.6250 - 303ms/epoch - 3ms/step\n",
      "Epoch 19/100\n",
      "93/93 - 0s - loss: 0.0121 - mse: 0.0242 - mae: 0.1084 - mape: 1788559.5000 - 312ms/epoch - 3ms/step\n",
      "Epoch 20/100\n",
      "93/93 - 0s - loss: 0.0120 - mse: 0.0241 - mae: 0.1086 - mape: 1775871.8750 - 314ms/epoch - 3ms/step\n",
      "Epoch 21/100\n",
      "93/93 - 0s - loss: 0.0121 - mse: 0.0243 - mae: 0.1095 - mape: 1844010.6250 - 314ms/epoch - 3ms/step\n",
      "Epoch 22/100\n",
      "93/93 - 0s - loss: 0.0119 - mse: 0.0238 - mae: 0.1022 - mape: 1592741.6250 - 324ms/epoch - 3ms/step\n",
      "Epoch 23/100\n",
      "93/93 - 0s - loss: 0.0119 - mse: 0.0238 - mae: 0.1098 - mape: 1797415.8750 - 326ms/epoch - 4ms/step\n",
      "Epoch 24/100\n",
      "93/93 - 0s - loss: 0.0118 - mse: 0.0236 - mae: 0.1038 - mape: 1597200.3750 - 324ms/epoch - 3ms/step\n",
      "Epoch 25/100\n",
      "93/93 - 0s - loss: 0.0118 - mse: 0.0237 - mae: 0.1057 - mape: 1622092.8750 - 327ms/epoch - 4ms/step\n",
      "Epoch 26/100\n",
      "93/93 - 0s - loss: 0.0117 - mse: 0.0234 - mae: 0.1079 - mape: 1749671.3750 - 324ms/epoch - 3ms/step\n",
      "Epoch 27/100\n",
      "93/93 - 0s - loss: 0.0116 - mse: 0.0232 - mae: 0.1064 - mape: 1753594.1250 - 321ms/epoch - 3ms/step\n",
      "Epoch 28/100\n",
      "93/93 - 0s - loss: 0.0116 - mse: 0.0232 - mae: 0.1038 - mape: 1723545.1250 - 335ms/epoch - 4ms/step\n",
      "Epoch 29/100\n",
      "93/93 - 0s - loss: 0.0115 - mse: 0.0230 - mae: 0.1073 - mape: 1618936.6250 - 322ms/epoch - 3ms/step\n",
      "Epoch 30/100\n",
      "93/93 - 0s - loss: 0.0115 - mse: 0.0231 - mae: 0.1057 - mape: 1642332.8750 - 326ms/epoch - 4ms/step\n",
      "Epoch 31/100\n",
      "93/93 - 0s - loss: 0.0115 - mse: 0.0230 - mae: 0.1016 - mape: 1694466.8750 - 314ms/epoch - 3ms/step\n",
      "Epoch 32/100\n",
      "93/93 - 0s - loss: 0.0114 - mse: 0.0228 - mae: 0.1034 - mape: 1586446.1250 - 322ms/epoch - 3ms/step\n",
      "Epoch 33/100\n",
      "93/93 - 0s - loss: 0.0113 - mse: 0.0226 - mae: 0.1035 - mape: 1643817.7500 - 328ms/epoch - 4ms/step\n",
      "Epoch 34/100\n",
      "93/93 - 0s - loss: 0.0113 - mse: 0.0226 - mae: 0.1062 - mape: 1664924.5000 - 303ms/epoch - 3ms/step\n",
      "Epoch 35/100\n",
      "93/93 - 0s - loss: 0.0113 - mse: 0.0226 - mae: 0.1053 - mape: 1550123.3750 - 283ms/epoch - 3ms/step\n",
      "Epoch 36/100\n",
      "93/93 - 0s - loss: 0.0115 - mse: 0.0230 - mae: 0.0998 - mape: 1430788.2500 - 275ms/epoch - 3ms/step\n",
      "Epoch 37/100\n",
      "93/93 - 0s - loss: 0.0112 - mse: 0.0224 - mae: 0.1065 - mape: 1806960.5000 - 281ms/epoch - 3ms/step\n",
      "Epoch 38/100\n",
      "93/93 - 0s - loss: 0.0112 - mse: 0.0224 - mae: 0.1014 - mape: 1536655.2500 - 281ms/epoch - 3ms/step\n",
      "Epoch 39/100\n",
      "93/93 - 0s - loss: 0.0111 - mse: 0.0222 - mae: 0.1037 - mape: 1625965.3750 - 283ms/epoch - 3ms/step\n",
      "Epoch 40/100\n",
      "93/93 - 0s - loss: 0.0111 - mse: 0.0222 - mae: 0.1038 - mape: 1590686.2500 - 282ms/epoch - 3ms/step\n",
      "Epoch 41/100\n",
      "93/93 - 0s - loss: 0.0110 - mse: 0.0220 - mae: 0.1017 - mape: 1509716.1250 - 275ms/epoch - 3ms/step\n",
      "Epoch 42/100\n",
      "93/93 - 0s - loss: 0.0111 - mse: 0.0221 - mae: 0.1017 - mape: 1499269.5000 - 274ms/epoch - 3ms/step\n",
      "Epoch 43/100\n",
      "93/93 - 0s - loss: 0.0112 - mse: 0.0225 - mae: 0.1066 - mape: 1600696.7500 - 273ms/epoch - 3ms/step\n",
      "Epoch 44/100\n",
      "93/93 - 0s - loss: 0.0112 - mse: 0.0224 - mae: 0.1047 - mape: 1547713.8750 - 283ms/epoch - 3ms/step\n",
      "Epoch 45/100\n",
      "93/93 - 0s - loss: 0.0110 - mse: 0.0219 - mae: 0.1017 - mape: 1530820.0000 - 283ms/epoch - 3ms/step\n",
      "Epoch 46/100\n",
      "93/93 - 0s - loss: 0.0109 - mse: 0.0218 - mae: 0.1024 - mape: 1532939.1250 - 296ms/epoch - 3ms/step\n",
      "Epoch 47/100\n",
      "93/93 - 0s - loss: 0.0109 - mse: 0.0219 - mae: 0.1018 - mape: 1445817.0000 - 323ms/epoch - 3ms/step\n",
      "Epoch 48/100\n",
      "93/93 - 0s - loss: 0.0109 - mse: 0.0218 - mae: 0.1019 - mape: 1523398.7500 - 325ms/epoch - 3ms/step\n",
      "Epoch 49/100\n",
      "93/93 - 0s - loss: 0.0109 - mse: 0.0218 - mae: 0.1009 - mape: 1537429.5000 - 313ms/epoch - 3ms/step\n",
      "Epoch 50/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0216 - mae: 0.1044 - mape: 1524113.3750 - 320ms/epoch - 3ms/step\n",
      "Epoch 51/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0216 - mae: 0.1009 - mape: 1436805.5000 - 302ms/epoch - 3ms/step\n",
      "Epoch 52/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0215 - mae: 0.1026 - mape: 1522211.5000 - 310ms/epoch - 3ms/step\n",
      "Epoch 53/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0217 - mae: 0.1032 - mape: 1519682.7500 - 317ms/epoch - 3ms/step\n",
      "Epoch 54/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0215 - mae: 0.1021 - mape: 1468466.6250 - 311ms/epoch - 3ms/step\n",
      "Epoch 55/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0217 - mae: 0.1043 - mape: 1369871.0000 - 304ms/epoch - 3ms/step\n",
      "Epoch 56/100\n",
      "93/93 - 0s - loss: 0.0107 - mse: 0.0215 - mae: 0.1035 - mape: 1452424.6250 - 303ms/epoch - 3ms/step\n",
      "Epoch 57/100\n",
      "93/93 - 0s - loss: 0.0108 - mse: 0.0215 - mae: 0.1029 - mape: 1402222.8750 - 290ms/epoch - 3ms/step\n",
      "Epoch 58/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0213 - mae: 0.1013 - mape: 1462650.5000 - 302ms/epoch - 3ms/step\n",
      "Epoch 59/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0213 - mae: 0.1027 - mape: 1441578.2500 - 314ms/epoch - 3ms/step\n",
      "Epoch 60/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0213 - mae: 0.1024 - mape: 1488526.6250 - 313ms/epoch - 3ms/step\n",
      "Epoch 61/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0211 - mae: 0.1025 - mape: 1388385.3750 - 322ms/epoch - 3ms/step\n",
      "Epoch 62/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0212 - mae: 0.1017 - mape: 1413219.1250 - 324ms/epoch - 3ms/step\n",
      "Epoch 63/100\n",
      "93/93 - 0s - loss: 0.0107 - mse: 0.0214 - mae: 0.1034 - mape: 1473346.3750 - 329ms/epoch - 4ms/step\n",
      "Epoch 64/100\n",
      "93/93 - 0s - loss: 0.0107 - mse: 0.0213 - mae: 0.1041 - mape: 1468474.5000 - 324ms/epoch - 3ms/step\n",
      "Epoch 65/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0211 - mae: 0.1029 - mape: 1460365.7500 - 316ms/epoch - 3ms/step\n",
      "Epoch 66/100\n",
      "93/93 - 0s - loss: 0.0107 - mse: 0.0214 - mae: 0.1023 - mape: 1437544.3750 - 310ms/epoch - 3ms/step\n",
      "Epoch 67/100\n",
      "93/93 - 0s - loss: 0.0105 - mse: 0.0210 - mae: 0.1014 - mape: 1381706.5000 - 327ms/epoch - 4ms/step\n",
      "Epoch 68/100\n",
      "93/93 - 0s - loss: 0.0105 - mse: 0.0211 - mae: 0.1022 - mape: 1348201.5000 - 324ms/epoch - 3ms/step\n",
      "Epoch 69/100\n",
      "93/93 - 0s - loss: 0.0106 - mse: 0.0211 - mae: 0.1033 - mape: 1399804.2500 - 312ms/epoch - 3ms/step\n",
      "Epoch 70/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0207 - mae: 0.1021 - mape: 1474550.7500 - 325ms/epoch - 3ms/step\n",
      "Epoch 71/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0208 - mae: 0.1017 - mape: 1402357.1250 - 317ms/epoch - 3ms/step\n",
      "Epoch 72/100\n",
      "93/93 - 0s - loss: 0.0105 - mse: 0.0210 - mae: 0.1013 - mape: 1330159.0000 - 324ms/epoch - 3ms/step\n",
      "Epoch 73/100\n",
      "93/93 - 0s - loss: 0.0105 - mse: 0.0209 - mae: 0.1024 - mape: 1478704.0000 - 327ms/epoch - 4ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "93/93 - 0s - loss: 0.0105 - mse: 0.0209 - mae: 0.1028 - mape: 1269563.0000 - 315ms/epoch - 3ms/step\n",
      "Epoch 75/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0208 - mae: 0.1032 - mape: 1521131.0000 - 314ms/epoch - 3ms/step\n",
      "Epoch 76/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0207 - mae: 0.1027 - mape: 1452821.3750 - 313ms/epoch - 3ms/step\n",
      "Epoch 77/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0208 - mae: 0.1031 - mape: 1443888.0000 - 314ms/epoch - 3ms/step\n",
      "Epoch 78/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0208 - mae: 0.1022 - mape: 1321556.5000 - 314ms/epoch - 3ms/step\n",
      "Epoch 79/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0208 - mae: 0.1015 - mape: 1451939.1250 - 295ms/epoch - 3ms/step\n",
      "Epoch 80/100\n",
      "93/93 - 0s - loss: 0.0104 - mse: 0.0207 - mae: 0.1030 - mape: 1395752.3750 - 313ms/epoch - 3ms/step\n",
      "Epoch 81/100\n",
      "93/93 - 0s - loss: 0.0103 - mse: 0.0205 - mae: 0.1014 - mape: 1298701.3750 - 303ms/epoch - 3ms/step\n",
      "Epoch 82/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0205 - mae: 0.1014 - mape: 1337145.8750 - 303ms/epoch - 3ms/step\n",
      "Epoch 83/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0204 - mae: 0.1037 - mape: 1467801.7500 - 314ms/epoch - 3ms/step\n",
      "Epoch 84/100\n",
      "93/93 - 0s - loss: 0.0101 - mse: 0.0202 - mae: 0.1022 - mape: 1420352.7500 - 296ms/epoch - 3ms/step\n",
      "Epoch 85/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0204 - mae: 0.1001 - mape: 1308899.2500 - 293ms/epoch - 3ms/step\n",
      "Epoch 86/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0204 - mae: 0.1017 - mape: 1366777.0000 - 305ms/epoch - 3ms/step\n",
      "Epoch 87/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0203 - mae: 0.1021 - mape: 1450904.7500 - 313ms/epoch - 3ms/step\n",
      "Epoch 88/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0203 - mae: 0.1038 - mape: 1488577.3750 - 312ms/epoch - 3ms/step\n",
      "Epoch 89/100\n",
      "93/93 - 0s - loss: 0.0101 - mse: 0.0202 - mae: 0.1012 - mape: 1374346.1250 - 314ms/epoch - 3ms/step\n",
      "Epoch 90/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0205 - mae: 0.1029 - mape: 1309918.6250 - 318ms/epoch - 3ms/step\n",
      "Epoch 91/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0205 - mae: 0.1026 - mape: 1326261.5000 - 316ms/epoch - 3ms/step\n",
      "Epoch 92/100\n",
      "93/93 - 0s - loss: 0.0102 - mse: 0.0204 - mae: 0.1039 - mape: 1320991.0000 - 314ms/epoch - 3ms/step\n",
      "Epoch 93/100\n",
      "93/93 - 0s - loss: 0.0100 - mse: 0.0200 - mae: 0.1005 - mape: 1295663.1250 - 314ms/epoch - 3ms/step\n",
      "Epoch 94/100\n",
      "93/93 - 0s - loss: 0.0101 - mse: 0.0201 - mae: 0.1023 - mape: 1345388.5000 - 313ms/epoch - 3ms/step\n",
      "Epoch 95/100\n",
      "93/93 - 0s - loss: 0.0100 - mse: 0.0199 - mae: 0.1013 - mape: 1380109.3750 - 312ms/epoch - 3ms/step\n",
      "Epoch 96/100\n",
      "93/93 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.1003 - mape: 1321975.0000 - 334ms/epoch - 4ms/step\n",
      "Epoch 97/100\n",
      "93/93 - 0s - loss: 0.0100 - mse: 0.0201 - mae: 0.1001 - mape: 1215881.8750 - 353ms/epoch - 4ms/step\n",
      "Epoch 98/100\n",
      "93/93 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.1019 - mape: 1424709.6250 - 353ms/epoch - 4ms/step\n",
      "Epoch 99/100\n",
      "93/93 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.1010 - mape: 1378358.1250 - 325ms/epoch - 3ms/step\n",
      "Epoch 100/100\n",
      "93/93 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.1017 - mape: 1377410.5000 - 313ms/epoch - 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2954da024a0>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 2 from reference but different loss function \n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(4, input_shape=(1, look_back)))\n",
    "model.add(Dense(1))\n",
    "# check https://medium.com/@tpyrt1069098/there-are-14-loss-functions-for-regression-tasks-you-know-several-of-them-5ecd5e77a599\n",
    "model.compile(loss='huber', optimizer='adam', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=1, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "bc05b218",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/250\n",
      "24/24 - 2s - loss: 0.0366 - mse: 0.0732 - mae: 0.2048 - mape: 343699.0312 - 2s/epoch - 76ms/step\n",
      "Epoch 2/250\n",
      "24/24 - 0s - loss: 0.0201 - mse: 0.0402 - mae: 0.1370 - mape: 1008580.8125 - 67ms/epoch - 3ms/step\n",
      "Epoch 3/250\n",
      "24/24 - 0s - loss: 0.0136 - mse: 0.0272 - mae: 0.1039 - mape: 1555595.5000 - 64ms/epoch - 3ms/step\n",
      "Epoch 4/250\n",
      "24/24 - 0s - loss: 0.0123 - mse: 0.0245 - mae: 0.1091 - mape: 1657082.1250 - 68ms/epoch - 3ms/step\n",
      "Epoch 5/250\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0243 - mae: 0.1077 - mape: 1658945.8750 - 59ms/epoch - 2ms/step\n",
      "Epoch 6/250\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0243 - mae: 0.1051 - mape: 1611633.2500 - 71ms/epoch - 3ms/step\n",
      "Epoch 7/250\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0243 - mae: 0.1092 - mape: 1734795.1250 - 73ms/epoch - 3ms/step\n",
      "Epoch 8/250\n",
      "24/24 - 0s - loss: 0.0120 - mse: 0.0239 - mae: 0.1087 - mape: 1744493.3750 - 63ms/epoch - 3ms/step\n",
      "Epoch 9/250\n",
      "24/24 - 0s - loss: 0.0119 - mse: 0.0239 - mae: 0.1063 - mape: 1664593.2500 - 73ms/epoch - 3ms/step\n",
      "Epoch 10/250\n",
      "24/24 - 0s - loss: 0.0119 - mse: 0.0238 - mae: 0.1091 - mape: 1768155.0000 - 71ms/epoch - 3ms/step\n",
      "Epoch 11/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0236 - mae: 0.1071 - mape: 1708428.3750 - 74ms/epoch - 3ms/step\n",
      "Epoch 12/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0236 - mae: 0.1052 - mape: 1655130.8750 - 75ms/epoch - 3ms/step\n",
      "Epoch 13/250\n",
      "24/24 - 0s - loss: 0.0117 - mse: 0.0234 - mae: 0.1074 - mape: 1699502.1250 - 74ms/epoch - 3ms/step\n",
      "Epoch 14/250\n",
      "24/24 - 0s - loss: 0.0116 - mse: 0.0231 - mae: 0.1047 - mape: 1673862.0000 - 74ms/epoch - 3ms/step\n",
      "Epoch 15/250\n",
      "24/24 - 0s - loss: 0.0116 - mse: 0.0232 - mae: 0.1038 - mape: 1598255.0000 - 71ms/epoch - 3ms/step\n",
      "Epoch 16/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0236 - mae: 0.1095 - mape: 1708468.1250 - 79ms/epoch - 3ms/step\n",
      "Epoch 17/250\n",
      "24/24 - 0s - loss: 0.0116 - mse: 0.0232 - mae: 0.1044 - mape: 1598486.3750 - 70ms/epoch - 3ms/step\n",
      "Epoch 18/250\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0227 - mae: 0.1054 - mape: 1635912.1250 - 74ms/epoch - 3ms/step\n",
      "Epoch 19/250\n",
      "24/24 - 0s - loss: 0.0114 - mse: 0.0227 - mae: 0.1050 - mape: 1654331.1250 - 69ms/epoch - 3ms/step\n",
      "Epoch 20/250\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0226 - mae: 0.1040 - mape: 1640633.2500 - 69ms/epoch - 3ms/step\n",
      "Epoch 21/250\n",
      "24/24 - 0s - loss: 0.0114 - mse: 0.0227 - mae: 0.1062 - mape: 1696230.2500 - 73ms/epoch - 3ms/step\n",
      "Epoch 22/250\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0226 - mae: 0.1031 - mape: 1536238.5000 - 73ms/epoch - 3ms/step\n",
      "Epoch 23/250\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0225 - mae: 0.1061 - mape: 1633308.8750 - 69ms/epoch - 3ms/step\n",
      "Epoch 24/250\n",
      "24/24 - 0s - loss: 0.0112 - mse: 0.0224 - mae: 0.1019 - mape: 1586330.6250 - 72ms/epoch - 3ms/step\n",
      "Epoch 25/250\n",
      "24/24 - 0s - loss: 0.0110 - mse: 0.0221 - mae: 0.1027 - mape: 1555390.6250 - 70ms/epoch - 3ms/step\n",
      "Epoch 26/250\n",
      "24/24 - 0s - loss: 0.0110 - mse: 0.0220 - mae: 0.1054 - mape: 1669654.5000 - 70ms/epoch - 3ms/step\n",
      "Epoch 27/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0218 - mae: 0.1038 - mape: 1628452.0000 - 71ms/epoch - 3ms/step\n",
      "Epoch 28/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0218 - mae: 0.1041 - mape: 1612986.5000 - 70ms/epoch - 3ms/step\n",
      "Epoch 29/250\n",
      "24/24 - 0s - loss: 0.0110 - mse: 0.0219 - mae: 0.1066 - mape: 1550510.6250 - 67ms/epoch - 3ms/step\n",
      "Epoch 30/250\n",
      "24/24 - 0s - loss: 0.0107 - mse: 0.0213 - mae: 0.1028 - mape: 1567755.3750 - 68ms/epoch - 3ms/step\n",
      "Epoch 31/250\n",
      "24/24 - 0s - loss: 0.0108 - mse: 0.0216 - mae: 0.1006 - mape: 1516033.3750 - 75ms/epoch - 3ms/step\n",
      "Epoch 32/250\n",
      "24/24 - 0s - loss: 0.0106 - mse: 0.0211 - mae: 0.1015 - mape: 1510180.2500 - 62ms/epoch - 3ms/step\n",
      "Epoch 33/250\n",
      "24/24 - 0s - loss: 0.0105 - mse: 0.0210 - mae: 0.1008 - mape: 1540450.7500 - 76ms/epoch - 3ms/step\n",
      "Epoch 34/250\n",
      "24/24 - 0s - loss: 0.0105 - mse: 0.0210 - mae: 0.1034 - mape: 1576215.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 35/250\n",
      "24/24 - 0s - loss: 0.0106 - mse: 0.0212 - mae: 0.1032 - mape: 1557476.5000 - 81ms/epoch - 3ms/step\n",
      "Epoch 36/250\n",
      "24/24 - 0s - loss: 0.0106 - mse: 0.0211 - mae: 0.0995 - mape: 1423073.3750 - 80ms/epoch - 3ms/step\n",
      "Epoch 37/250\n",
      "24/24 - 0s - loss: 0.0105 - mse: 0.0211 - mae: 0.1041 - mape: 1575110.7500 - 81ms/epoch - 3ms/step\n",
      "Epoch 38/250\n",
      "24/24 - 0s - loss: 0.0103 - mse: 0.0206 - mae: 0.0992 - mape: 1432849.2500 - 71ms/epoch - 3ms/step\n",
      "Epoch 39/250\n",
      "24/24 - 0s - loss: 0.0101 - mse: 0.0203 - mae: 0.1005 - mape: 1530102.2500 - 72ms/epoch - 3ms/step\n",
      "Epoch 40/250\n",
      "24/24 - 0s - loss: 0.0100 - mse: 0.0201 - mae: 0.1000 - mape: 1508925.2500 - 73ms/epoch - 3ms/step\n",
      "Epoch 41/250\n",
      "24/24 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.0986 - mape: 1466947.2500 - 71ms/epoch - 3ms/step\n",
      "Epoch 42/250\n",
      "24/24 - 0s - loss: 0.0102 - mse: 0.0203 - mae: 0.0986 - mape: 1409183.2500 - 60ms/epoch - 3ms/step\n",
      "Epoch 43/250\n",
      "24/24 - 0s - loss: 0.0099 - mse: 0.0198 - mae: 0.0988 - mape: 1428026.2500 - 61ms/epoch - 3ms/step\n",
      "Epoch 44/250\n",
      "24/24 - 0s - loss: 0.0097 - mse: 0.0195 - mae: 0.0988 - mape: 1455299.1250 - 69ms/epoch - 3ms/step\n",
      "Epoch 45/250\n",
      "24/24 - 0s - loss: 0.0096 - mse: 0.0192 - mae: 0.0978 - mape: 1438758.6250 - 63ms/epoch - 3ms/step\n",
      "Epoch 46/250\n",
      "24/24 - 0s - loss: 0.0096 - mse: 0.0192 - mae: 0.0985 - mape: 1411385.8750 - 67ms/epoch - 3ms/step\n",
      "Epoch 47/250\n",
      "24/24 - 0s - loss: 0.0096 - mse: 0.0192 - mae: 0.0989 - mape: 1441774.2500 - 67ms/epoch - 3ms/step\n",
      "Epoch 48/250\n",
      "24/24 - 0s - loss: 0.0094 - mse: 0.0188 - mae: 0.0965 - mape: 1402469.6250 - 68ms/epoch - 3ms/step\n",
      "Epoch 49/250\n",
      "24/24 - 0s - loss: 0.0095 - mse: 0.0191 - mae: 0.0952 - mape: 1333288.7500 - 75ms/epoch - 3ms/step\n",
      "Epoch 50/250\n",
      "24/24 - 0s - loss: 0.0093 - mse: 0.0187 - mae: 0.0992 - mape: 1399109.3750 - 71ms/epoch - 3ms/step\n",
      "Epoch 51/250\n",
      "24/24 - 0s - loss: 0.0092 - mse: 0.0185 - mae: 0.0957 - mape: 1304532.2500 - 73ms/epoch - 3ms/step\n",
      "Epoch 52/250\n",
      "24/24 - 0s - loss: 0.0091 - mse: 0.0181 - mae: 0.0960 - mape: 1384155.6250 - 67ms/epoch - 3ms/step\n",
      "Epoch 53/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0180 - mae: 0.0957 - mape: 1367945.2500 - 71ms/epoch - 3ms/step\n",
      "Epoch 54/250\n",
      "24/24 - 0s - loss: 0.0089 - mse: 0.0179 - mae: 0.0950 - mape: 1308880.7500 - 76ms/epoch - 3ms/step\n",
      "Epoch 55/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0181 - mae: 0.0978 - mape: 1295836.6250 - 70ms/epoch - 3ms/step\n",
      "Epoch 56/250\n",
      "24/24 - 0s - loss: 0.0091 - mse: 0.0183 - mae: 0.1015 - mape: 1424273.7500 - 70ms/epoch - 3ms/step\n",
      "Epoch 57/250\n",
      "24/24 - 0s - loss: 0.0089 - mse: 0.0179 - mae: 0.0959 - mape: 1342337.0000 - 67ms/epoch - 3ms/step\n",
      "Epoch 58/250\n",
      "24/24 - 0s - loss: 0.0087 - mse: 0.0174 - mae: 0.0934 - mape: 1268228.0000 - 72ms/epoch - 3ms/step\n",
      "Epoch 59/250\n",
      "24/24 - 0s - loss: 0.0085 - mse: 0.0171 - mae: 0.0947 - mape: 1301394.8750 - 70ms/epoch - 3ms/step\n",
      "Epoch 60/250\n",
      "24/24 - 0s - loss: 0.0085 - mse: 0.0169 - mae: 0.0946 - mape: 1344918.3750 - 73ms/epoch - 3ms/step\n",
      "Epoch 61/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0165 - mae: 0.0937 - mape: 1230562.6250 - 68ms/epoch - 3ms/step\n",
      "Epoch 62/250\n",
      "24/24 - 0s - loss: 0.0084 - mse: 0.0167 - mae: 0.0909 - mape: 1192036.3750 - 71ms/epoch - 3ms/step\n",
      "Epoch 63/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0165 - mae: 0.0941 - mape: 1254987.7500 - 67ms/epoch - 3ms/step\n",
      "Epoch 64/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0165 - mae: 0.0935 - mape: 1285604.1250 - 70ms/epoch - 3ms/step\n",
      "Epoch 65/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0160 - mae: 0.0926 - mape: 1285274.5000 - 82ms/epoch - 3ms/step\n",
      "Epoch 66/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0167 - mae: 0.0923 - mape: 1314468.3750 - 71ms/epoch - 3ms/step\n",
      "Epoch 67/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0160 - mae: 0.0901 - mape: 1234580.7500 - 70ms/epoch - 3ms/step\n",
      "Epoch 68/250\n",
      "24/24 - 0s - loss: 0.0079 - mse: 0.0157 - mae: 0.0897 - mape: 1125828.0000 - 72ms/epoch - 3ms/step\n",
      "Epoch 69/250\n",
      "24/24 - 0s - loss: 0.0078 - mse: 0.0156 - mae: 0.0906 - mape: 1175884.3750 - 70ms/epoch - 3ms/step\n",
      "Epoch 70/250\n",
      "24/24 - 0s - loss: 0.0078 - mse: 0.0157 - mae: 0.0909 - mape: 1205764.0000 - 70ms/epoch - 3ms/step\n",
      "Epoch 71/250\n",
      "24/24 - 0s - loss: 0.0077 - mse: 0.0153 - mae: 0.0893 - mape: 1190785.7500 - 70ms/epoch - 3ms/step\n",
      "Epoch 72/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0153 - mae: 0.0864 - mape: 1064055.8750 - 75ms/epoch - 3ms/step\n",
      "Epoch 73/250\n",
      "24/24 - 0s - loss: 0.0077 - mse: 0.0154 - mae: 0.0903 - mape: 1184660.1250 - 70ms/epoch - 3ms/step\n",
      "Epoch 74/250\n",
      "24/24 - 0s - loss: 0.0078 - mse: 0.0155 - mae: 0.0914 - mape: 1126628.2500 - 70ms/epoch - 3ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 75/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0151 - mae: 0.0880 - mape: 1115463.0000 - 76ms/epoch - 3ms/step\n",
      "Epoch 76/250\n",
      "24/24 - 0s - loss: 0.0074 - mse: 0.0148 - mae: 0.0874 - mape: 1120016.3750 - 80ms/epoch - 3ms/step\n",
      "Epoch 77/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0149 - mae: 0.0886 - mape: 1093787.8750 - 81ms/epoch - 3ms/step\n",
      "Epoch 78/250\n",
      "24/24 - 0s - loss: 0.0073 - mse: 0.0146 - mae: 0.0868 - mape: 1081973.6250 - 72ms/epoch - 3ms/step\n",
      "Epoch 79/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0144 - mae: 0.0871 - mape: 1106937.7500 - 71ms/epoch - 3ms/step\n",
      "Epoch 80/250\n",
      "24/24 - 0s - loss: 0.0073 - mse: 0.0145 - mae: 0.0860 - mape: 1041477.9375 - 75ms/epoch - 3ms/step\n",
      "Epoch 81/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0144 - mae: 0.0850 - mape: 1015425.4375 - 67ms/epoch - 3ms/step\n",
      "Epoch 82/250\n",
      "24/24 - 0s - loss: 0.0071 - mse: 0.0142 - mae: 0.0861 - mape: 1049847.5000 - 75ms/epoch - 3ms/step\n",
      "Epoch 83/250\n",
      "24/24 - 0s - loss: 0.0071 - mse: 0.0142 - mae: 0.0878 - mape: 1110248.7500 - 72ms/epoch - 3ms/step\n",
      "Epoch 84/250\n",
      "24/24 - 0s - loss: 0.0071 - mse: 0.0141 - mae: 0.0865 - mape: 1110588.8750 - 79ms/epoch - 3ms/step\n",
      "Epoch 85/250\n",
      "24/24 - 0s - loss: 0.0070 - mse: 0.0139 - mae: 0.0841 - mape: 951561.5625 - 91ms/epoch - 4ms/step\n",
      "Epoch 86/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0137 - mae: 0.0856 - mape: 996473.3125 - 81ms/epoch - 3ms/step\n",
      "Epoch 87/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0137 - mae: 0.0856 - mape: 1040261.9375 - 87ms/epoch - 4ms/step\n",
      "Epoch 88/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0139 - mae: 0.0864 - mape: 1038086.5625 - 93ms/epoch - 4ms/step\n",
      "Epoch 89/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0136 - mae: 0.0819 - mape: 946499.4375 - 89ms/epoch - 4ms/step\n",
      "Epoch 90/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0137 - mae: 0.0845 - mape: 980251.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 91/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0135 - mae: 0.0855 - mape: 1010706.8125 - 90ms/epoch - 4ms/step\n",
      "Epoch 92/250\n",
      "24/24 - 0s - loss: 0.0066 - mse: 0.0132 - mae: 0.0841 - mape: 990948.6250 - 93ms/epoch - 4ms/step\n",
      "Epoch 93/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0136 - mae: 0.0826 - mape: 852315.8125 - 93ms/epoch - 4ms/step\n",
      "Epoch 94/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0133 - mae: 0.0861 - mape: 1078365.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 95/250\n",
      "24/24 - 0s - loss: 0.0066 - mse: 0.0132 - mae: 0.0829 - mape: 969463.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 96/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0128 - mae: 0.0818 - mape: 931406.1875 - 89ms/epoch - 4ms/step\n",
      "Epoch 97/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0130 - mae: 0.0830 - mape: 977462.1250 - 90ms/epoch - 4ms/step\n",
      "Epoch 98/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0127 - mae: 0.0818 - mape: 942356.3125 - 90ms/epoch - 4ms/step\n",
      "Epoch 99/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0127 - mae: 0.0821 - mape: 948304.7500 - 84ms/epoch - 3ms/step\n",
      "Epoch 100/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0129 - mae: 0.0835 - mape: 969432.9375 - 92ms/epoch - 4ms/step\n",
      "Epoch 101/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0129 - mae: 0.0823 - mape: 949802.6875 - 85ms/epoch - 4ms/step\n",
      "Epoch 102/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0127 - mae: 0.0815 - mape: 802294.9375 - 84ms/epoch - 3ms/step\n",
      "Epoch 103/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0129 - mae: 0.0825 - mape: 919954.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 104/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0125 - mae: 0.0830 - mape: 955396.8750 - 91ms/epoch - 4ms/step\n",
      "Epoch 105/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0126 - mae: 0.0813 - mape: 812739.9375 - 91ms/epoch - 4ms/step\n",
      "Epoch 106/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0125 - mae: 0.0829 - mape: 887369.1875 - 90ms/epoch - 4ms/step\n",
      "Epoch 107/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0125 - mae: 0.0810 - mape: 817555.9375 - 92ms/epoch - 4ms/step\n",
      "Epoch 108/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0124 - mae: 0.0799 - mape: 711204.8125 - 90ms/epoch - 4ms/step\n",
      "Epoch 109/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0123 - mae: 0.0827 - mape: 957248.5000 - 83ms/epoch - 3ms/step\n",
      "Epoch 110/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0135 - mae: 0.0829 - mape: 785942.6875 - 92ms/epoch - 4ms/step\n",
      "Epoch 111/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0124 - mae: 0.0814 - mape: 874001.7500 - 84ms/epoch - 3ms/step\n",
      "Epoch 112/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0121 - mae: 0.0809 - mape: 838699.9375 - 88ms/epoch - 4ms/step\n",
      "Epoch 113/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0124 - mae: 0.0814 - mape: 830140.1875 - 91ms/epoch - 4ms/step\n",
      "Epoch 114/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0122 - mae: 0.0813 - mape: 900382.8125 - 91ms/epoch - 4ms/step\n",
      "Epoch 115/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0121 - mae: 0.0808 - mape: 829876.1250 - 104ms/epoch - 4ms/step\n",
      "Epoch 116/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0121 - mae: 0.0819 - mape: 965327.5625 - 91ms/epoch - 4ms/step\n",
      "Epoch 117/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0123 - mae: 0.0805 - mape: 760654.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 118/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0119 - mae: 0.0817 - mape: 920273.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 119/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0121 - mae: 0.0819 - mape: 890042.7500 - 87ms/epoch - 4ms/step\n",
      "Epoch 120/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0129 - mae: 0.0844 - mape: 829479.5000 - 88ms/epoch - 4ms/step\n",
      "Epoch 121/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0121 - mae: 0.0802 - mape: 718194.3750 - 88ms/epoch - 4ms/step\n",
      "Epoch 122/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0117 - mae: 0.0800 - mape: 836552.6250 - 87ms/epoch - 4ms/step\n",
      "Epoch 123/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0124 - mae: 0.0817 - mape: 871424.7500 - 85ms/epoch - 4ms/step\n",
      "Epoch 124/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0124 - mae: 0.0838 - mape: 935132.5000 - 89ms/epoch - 4ms/step\n",
      "Epoch 125/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0118 - mae: 0.0801 - mape: 764882.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 126/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0791 - mape: 789930.4375 - 92ms/epoch - 4ms/step\n",
      "Epoch 127/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0121 - mae: 0.0815 - mape: 739666.1250 - 83ms/epoch - 3ms/step\n",
      "Epoch 128/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0123 - mae: 0.0841 - mape: 868709.9375 - 81ms/epoch - 3ms/step\n",
      "Epoch 129/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0127 - mae: 0.0852 - mape: 965493.7500 - 86ms/epoch - 4ms/step\n",
      "Epoch 130/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0117 - mae: 0.0777 - mape: 723428.1875 - 85ms/epoch - 4ms/step\n",
      "Epoch 131/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0786 - mape: 784684.5625 - 83ms/epoch - 3ms/step\n",
      "Epoch 132/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0782 - mape: 715610.8750 - 92ms/epoch - 4ms/step\n",
      "Epoch 133/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0796 - mape: 818043.0000 - 83ms/epoch - 3ms/step\n",
      "Epoch 134/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0115 - mae: 0.0794 - mape: 804908.3125 - 87ms/epoch - 4ms/step\n",
      "Epoch 135/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0117 - mae: 0.0812 - mape: 941948.5625 - 86ms/epoch - 4ms/step\n",
      "Epoch 136/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0126 - mae: 0.0828 - mape: 686151.2500 - 83ms/epoch - 3ms/step\n",
      "Epoch 137/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0119 - mae: 0.0798 - mape: 701851.5000 - 87ms/epoch - 4ms/step\n",
      "Epoch 138/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0792 - mape: 854717.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 139/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0115 - mae: 0.0791 - mape: 820601.8750 - 83ms/epoch - 3ms/step\n",
      "Epoch 140/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0114 - mae: 0.0790 - mape: 784006.1875 - 92ms/epoch - 4ms/step\n",
      "Epoch 141/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0114 - mae: 0.0786 - mape: 837060.1250 - 84ms/epoch - 4ms/step\n",
      "Epoch 142/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0775 - mape: 825887.1250 - 88ms/epoch - 4ms/step\n",
      "Epoch 143/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0799 - mape: 805323.6875 - 82ms/epoch - 3ms/step\n",
      "Epoch 144/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0783 - mape: 781693.6875 - 92ms/epoch - 4ms/step\n",
      "Epoch 145/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0114 - mae: 0.0786 - mape: 783907.2500 - 103ms/epoch - 4ms/step\n",
      "Epoch 146/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0118 - mae: 0.0804 - mape: 798054.8750 - 84ms/epoch - 3ms/step\n",
      "Epoch 147/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0116 - mae: 0.0770 - mape: 775632.2500 - 88ms/epoch - 4ms/step\n",
      "Epoch 148/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0113 - mae: 0.0791 - mape: 747665.1875 - 91ms/epoch - 4ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 149/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0117 - mae: 0.0803 - mape: 792530.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 150/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0114 - mae: 0.0790 - mape: 822623.5000 - 88ms/epoch - 4ms/step\n",
      "Epoch 151/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0775 - mape: 717716.6250 - 88ms/epoch - 4ms/step\n",
      "Epoch 152/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0115 - mae: 0.0794 - mape: 682589.0625 - 89ms/epoch - 4ms/step\n",
      "Epoch 153/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0779 - mape: 816207.2500 - 100ms/epoch - 4ms/step\n",
      "Epoch 154/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0783 - mape: 821592.1875 - 87ms/epoch - 4ms/step\n",
      "Epoch 155/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0113 - mae: 0.0788 - mape: 794312.3750 - 91ms/epoch - 4ms/step\n",
      "Epoch 156/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0780 - mape: 862491.8750 - 85ms/epoch - 4ms/step\n",
      "Epoch 157/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0788 - mape: 742172.5625 - 94ms/epoch - 4ms/step\n",
      "Epoch 158/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0115 - mae: 0.0776 - mape: 664591.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 159/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0795 - mape: 837207.8125 - 84ms/epoch - 3ms/step\n",
      "Epoch 160/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0787 - mape: 922933.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 161/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0776 - mape: 719976.5000 - 89ms/epoch - 4ms/step\n",
      "Epoch 162/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0784 - mape: 807642.8125 - 92ms/epoch - 4ms/step\n",
      "Epoch 163/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0774 - mape: 918928.4375 - 85ms/epoch - 4ms/step\n",
      "Epoch 164/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0777 - mape: 816224.8750 - 89ms/epoch - 4ms/step\n",
      "Epoch 165/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0767 - mape: 702104.5000 - 91ms/epoch - 4ms/step\n",
      "Epoch 166/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0111 - mae: 0.0778 - mape: 747710.1250 - 84ms/epoch - 3ms/step\n",
      "Epoch 167/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0108 - mae: 0.0761 - mape: 719692.0625 - 88ms/epoch - 4ms/step\n",
      "Epoch 168/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0108 - mae: 0.0755 - mape: 741200.6875 - 84ms/epoch - 4ms/step\n",
      "Epoch 169/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0768 - mape: 751027.9375 - 91ms/epoch - 4ms/step\n",
      "Epoch 170/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0109 - mae: 0.0766 - mape: 842207.3125 - 91ms/epoch - 4ms/step\n",
      "Epoch 171/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0108 - mae: 0.0762 - mape: 684357.8125 - 88ms/epoch - 4ms/step\n",
      "Epoch 172/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0759 - mape: 781253.5625 - 83ms/epoch - 3ms/step\n",
      "Epoch 173/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0752 - mape: 798161.8125 - 91ms/epoch - 4ms/step\n",
      "Epoch 174/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0752 - mape: 658721.5000 - 84ms/epoch - 3ms/step\n",
      "Epoch 175/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0107 - mae: 0.0746 - mape: 716920.6875 - 91ms/epoch - 4ms/step\n",
      "Epoch 176/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0773 - mape: 649157.1250 - 81ms/epoch - 3ms/step\n",
      "Epoch 177/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0105 - mae: 0.0753 - mape: 732541.3125 - 88ms/epoch - 4ms/step\n",
      "Epoch 178/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0105 - mae: 0.0740 - mape: 672232.7500 - 92ms/epoch - 4ms/step\n",
      "Epoch 179/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0742 - mape: 719596.2500 - 89ms/epoch - 4ms/step\n",
      "Epoch 180/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0745 - mape: 683937.5625 - 91ms/epoch - 4ms/step\n",
      "Epoch 181/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0743 - mape: 771207.5625 - 82ms/epoch - 3ms/step\n",
      "Epoch 182/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0758 - mape: 689584.8125 - 88ms/epoch - 4ms/step\n",
      "Epoch 183/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0103 - mae: 0.0741 - mape: 706795.6250 - 93ms/epoch - 4ms/step\n",
      "Epoch 184/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0105 - mae: 0.0744 - mape: 753167.5625 - 85ms/epoch - 4ms/step\n",
      "Epoch 185/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0762 - mape: 660756.3750 - 86ms/epoch - 4ms/step\n",
      "Epoch 186/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0105 - mae: 0.0749 - mape: 675434.5625 - 92ms/epoch - 4ms/step\n",
      "Epoch 187/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0103 - mae: 0.0741 - mape: 735610.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 188/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0103 - mae: 0.0741 - mape: 796710.6250 - 87ms/epoch - 4ms/step\n",
      "Epoch 189/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0759 - mape: 748962.0625 - 90ms/epoch - 4ms/step\n",
      "Epoch 190/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0743 - mape: 718362.3750 - 85ms/epoch - 4ms/step\n",
      "Epoch 191/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0754 - mape: 715538.6250 - 91ms/epoch - 4ms/step\n",
      "Epoch 192/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0110 - mae: 0.0802 - mape: 753984.6875 - 85ms/epoch - 4ms/step\n",
      "Epoch 193/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0105 - mae: 0.0759 - mape: 649394.6875 - 91ms/epoch - 4ms/step\n",
      "Epoch 194/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0105 - mae: 0.0731 - mape: 864667.8125 - 90ms/epoch - 4ms/step\n",
      "Epoch 195/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0102 - mae: 0.0730 - mape: 724125.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 196/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0102 - mae: 0.0740 - mape: 728917.0000 - 84ms/epoch - 3ms/step\n",
      "Epoch 197/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0102 - mae: 0.0742 - mape: 742663.8125 - 100ms/epoch - 4ms/step\n",
      "Epoch 198/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0102 - mae: 0.0746 - mape: 738058.4375 - 84ms/epoch - 4ms/step\n",
      "Epoch 199/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0109 - mae: 0.0791 - mape: 916618.5625 - 82ms/epoch - 3ms/step\n",
      "Epoch 200/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0105 - mae: 0.0771 - mape: 649403.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 201/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0103 - mae: 0.0752 - mape: 726638.3750 - 84ms/epoch - 3ms/step\n",
      "Epoch 202/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0099 - mae: 0.0737 - mape: 694373.8750 - 92ms/epoch - 4ms/step\n",
      "Epoch 203/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0740 - mape: 842369.7500 - 84ms/epoch - 3ms/step\n",
      "Epoch 204/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0112 - mae: 0.0788 - mape: 910032.9375 - 85ms/epoch - 4ms/step\n",
      "Epoch 205/250\n",
      "24/24 - 0s - loss: 0.0053 - mse: 0.0106 - mae: 0.0747 - mape: 707609.0625 - 88ms/epoch - 4ms/step\n",
      "Epoch 206/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0101 - mae: 0.0735 - mape: 667671.1875 - 85ms/epoch - 4ms/step\n",
      "Epoch 207/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0732 - mape: 651312.7500 - 82ms/epoch - 3ms/step\n",
      "Epoch 208/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0100 - mae: 0.0735 - mape: 738670.5625 - 82ms/epoch - 3ms/step\n",
      "Epoch 209/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0101 - mae: 0.0727 - mape: 682384.5000 - 88ms/epoch - 4ms/step\n",
      "Epoch 210/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0099 - mae: 0.0733 - mape: 622290.3125 - 91ms/epoch - 4ms/step\n",
      "Epoch 211/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0102 - mae: 0.0744 - mape: 808240.2500 - 83ms/epoch - 3ms/step\n",
      "Epoch 212/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0729 - mape: 702845.3750 - 87ms/epoch - 4ms/step\n",
      "Epoch 213/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0726 - mape: 716558.7500 - 81ms/epoch - 3ms/step\n",
      "Epoch 214/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0727 - mape: 690013.5625 - 86ms/epoch - 4ms/step\n",
      "Epoch 215/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0729 - mape: 802814.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 216/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0097 - mae: 0.0731 - mape: 661460.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 217/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0735 - mape: 688177.8750 - 82ms/epoch - 3ms/step\n",
      "Epoch 218/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0097 - mae: 0.0717 - mape: 617196.3750 - 87ms/epoch - 4ms/step\n",
      "Epoch 219/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0100 - mae: 0.0739 - mape: 803371.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 220/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0100 - mae: 0.0743 - mape: 750748.1875 - 92ms/epoch - 4ms/step\n",
      "Epoch 221/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0720 - mape: 729056.7500 - 81ms/epoch - 3ms/step\n",
      "Epoch 222/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0097 - mae: 0.0721 - mape: 663319.0000 - 91ms/epoch - 4ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 223/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0095 - mae: 0.0719 - mape: 628506.6250 - 86ms/epoch - 4ms/step\n",
      "Epoch 224/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0104 - mae: 0.0758 - mape: 779210.4375 - 83ms/epoch - 3ms/step\n",
      "Epoch 225/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0713 - mape: 699358.1250 - 82ms/epoch - 3ms/step\n",
      "Epoch 226/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0719 - mape: 726831.8125 - 82ms/epoch - 3ms/step\n",
      "Epoch 227/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0733 - mape: 635151.6250 - 83ms/epoch - 3ms/step\n",
      "Epoch 228/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0716 - mape: 602319.5000 - 91ms/epoch - 4ms/step\n",
      "Epoch 229/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0100 - mae: 0.0736 - mape: 682804.6250 - 85ms/epoch - 4ms/step\n",
      "Epoch 230/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0739 - mape: 638980.6250 - 92ms/epoch - 4ms/step\n",
      "Epoch 231/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0095 - mae: 0.0711 - mape: 623614.5625 - 86ms/epoch - 4ms/step\n",
      "Epoch 232/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0097 - mae: 0.0736 - mape: 640402.6875 - 91ms/epoch - 4ms/step\n",
      "Epoch 233/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0734 - mape: 680941.0625 - 83ms/epoch - 3ms/step\n",
      "Epoch 234/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0731 - mape: 755647.0625 - 101ms/epoch - 4ms/step\n",
      "Epoch 235/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0093 - mae: 0.0714 - mape: 771723.1875 - 82ms/epoch - 3ms/step\n",
      "Epoch 236/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0098 - mae: 0.0739 - mape: 729226.3125 - 92ms/epoch - 4ms/step\n",
      "Epoch 237/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0091 - mae: 0.0702 - mape: 695595.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 238/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0093 - mae: 0.0711 - mape: 729685.1875 - 92ms/epoch - 4ms/step\n",
      "Epoch 239/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0719 - mape: 660710.9375 - 83ms/epoch - 3ms/step\n",
      "Epoch 240/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0092 - mae: 0.0704 - mape: 687764.2500 - 81ms/epoch - 3ms/step\n",
      "Epoch 241/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0099 - mae: 0.0729 - mape: 687458.8750 - 91ms/epoch - 4ms/step\n",
      "Epoch 242/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0093 - mae: 0.0704 - mape: 589475.1875 - 91ms/epoch - 4ms/step\n",
      "Epoch 243/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0093 - mae: 0.0708 - mape: 753296.0625 - 92ms/epoch - 4ms/step\n",
      "Epoch 244/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0091 - mae: 0.0704 - mape: 642024.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 245/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0095 - mae: 0.0702 - mape: 677159.8125 - 92ms/epoch - 4ms/step\n",
      "Epoch 246/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0096 - mae: 0.0732 - mape: 650129.7500 - 81ms/epoch - 3ms/step\n",
      "Epoch 247/250\n",
      "24/24 - 0s - loss: 0.0045 - mse: 0.0090 - mae: 0.0690 - mape: 709046.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 248/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0092 - mae: 0.0698 - mape: 741011.1875 - 91ms/epoch - 4ms/step\n",
      "Epoch 249/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0091 - mae: 0.0698 - mape: 703606.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 250/250\n",
      "24/24 - 0s - loss: 0.0045 - mse: 0.0091 - mae: 0.0710 - mape: 872151.5000 - 82ms/epoch - 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x29586e4c7f0>"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 3 best parameters as proposed from grid search\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(24, activation='relu', input_shape=(1, look_back)))\n",
    "# TBD add second dense layer with more units?\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='huber', optimizer='adam', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=250, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "5aac6ff7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/250\n",
      "24/24 - 3s - loss: 0.0909 - mse: 0.0909 - mae: 0.2389 - mape: 183147.3906 - 3s/epoch - 143ms/step\n",
      "Epoch 2/250\n",
      "24/24 - 0s - loss: 0.0515 - mse: 0.0515 - mae: 0.1648 - mape: 1204517.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 3/250\n",
      "24/24 - 0s - loss: 0.0283 - mse: 0.0283 - mae: 0.1245 - mape: 2140684.5000 - 93ms/epoch - 4ms/step\n",
      "Epoch 4/250\n",
      "24/24 - 0s - loss: 0.0243 - mse: 0.0243 - mae: 0.1121 - mape: 2020238.5000 - 92ms/epoch - 4ms/step\n",
      "Epoch 5/250\n",
      "24/24 - 0s - loss: 0.0257 - mse: 0.0257 - mae: 0.1115 - mape: 2086203.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 6/250\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1131 - mape: 1639316.6250 - 94ms/epoch - 4ms/step\n",
      "Epoch 7/250\n",
      "24/24 - 0s - loss: 0.0272 - mse: 0.0272 - mae: 0.1197 - mape: 2012722.3750 - 96ms/epoch - 4ms/step\n",
      "Epoch 8/250\n",
      "24/24 - 0s - loss: 0.0254 - mse: 0.0254 - mae: 0.1177 - mape: 2208213.2500 - 85ms/epoch - 4ms/step\n",
      "Epoch 9/250\n",
      "24/24 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1097 - mape: 1525343.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 10/250\n",
      "24/24 - 0s - loss: 0.0245 - mse: 0.0245 - mae: 0.1208 - mape: 1892354.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 11/250\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1180 - mape: 1810375.6250 - 85ms/epoch - 4ms/step\n",
      "Epoch 12/250\n",
      "24/24 - 0s - loss: 0.0241 - mse: 0.0241 - mae: 0.1126 - mape: 1766779.8750 - 91ms/epoch - 4ms/step\n",
      "Epoch 13/250\n",
      "24/24 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1158 - mape: 2122957.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 14/250\n",
      "24/24 - 0s - loss: 0.0255 - mse: 0.0255 - mae: 0.1141 - mape: 1586629.3750 - 103ms/epoch - 4ms/step\n",
      "Epoch 15/250\n",
      "24/24 - 0s - loss: 0.0241 - mse: 0.0241 - mae: 0.1071 - mape: 1816333.2500 - 100ms/epoch - 4ms/step\n",
      "Epoch 16/250\n",
      "24/24 - 0s - loss: 0.0251 - mse: 0.0251 - mae: 0.1172 - mape: 1687400.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 17/250\n",
      "24/24 - 0s - loss: 0.0248 - mse: 0.0248 - mae: 0.1073 - mape: 1437697.6250 - 84ms/epoch - 3ms/step\n",
      "Epoch 18/250\n",
      "24/24 - 0s - loss: 0.0247 - mse: 0.0247 - mae: 0.1132 - mape: 1875529.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 19/250\n",
      "24/24 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1074 - mape: 1433727.6250 - 91ms/epoch - 4ms/step\n",
      "Epoch 20/250\n",
      "24/24 - 0s - loss: 0.0215 - mse: 0.0215 - mae: 0.1056 - mape: 1680513.7500 - 92ms/epoch - 4ms/step\n",
      "Epoch 21/250\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1092 - mape: 1594859.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 22/250\n",
      "24/24 - 0s - loss: 0.0246 - mse: 0.0246 - mae: 0.1101 - mape: 1499324.5000 - 84ms/epoch - 3ms/step\n",
      "Epoch 23/250\n",
      "24/24 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1126 - mape: 1719586.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 24/250\n",
      "24/24 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1052 - mape: 1360639.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 25/250\n",
      "24/24 - 0s - loss: 0.0225 - mse: 0.0225 - mae: 0.1056 - mape: 1555039.8750 - 91ms/epoch - 4ms/step\n",
      "Epoch 26/250\n",
      "24/24 - 0s - loss: 0.0228 - mse: 0.0228 - mae: 0.1108 - mape: 1544127.6250 - 92ms/epoch - 4ms/step\n",
      "Epoch 27/250\n",
      "24/24 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1041 - mape: 1395939.5000 - 84ms/epoch - 3ms/step\n",
      "Epoch 28/250\n",
      "24/24 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1030 - mape: 1352196.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 29/250\n",
      "24/24 - 0s - loss: 0.0225 - mse: 0.0225 - mae: 0.1135 - mape: 1532473.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 30/250\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1083 - mape: 1053667.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 31/250\n",
      "24/24 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.0985 - mape: 1269299.0000 - 92ms/epoch - 4ms/step\n",
      "Epoch 32/250\n",
      "24/24 - 0s - loss: 0.0179 - mse: 0.0179 - mae: 0.0939 - mape: 1120319.8750 - 93ms/epoch - 4ms/step\n",
      "Epoch 33/250\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1025 - mape: 1441371.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 34/250\n",
      "24/24 - 0s - loss: 0.0184 - mse: 0.0184 - mae: 0.1054 - mape: 1071405.3750 - 93ms/epoch - 4ms/step\n",
      "Epoch 35/250\n",
      "24/24 - 0s - loss: 0.0200 - mse: 0.0200 - mae: 0.0978 - mape: 1334157.8750 - 91ms/epoch - 4ms/step\n",
      "Epoch 36/250\n",
      "24/24 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.0989 - mape: 1223229.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 37/250\n",
      "24/24 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.1036 - mape: 1557605.5000 - 91ms/epoch - 4ms/step\n",
      "Epoch 38/250\n",
      "24/24 - 0s - loss: 0.0193 - mse: 0.0193 - mae: 0.0944 - mape: 1260385.3750 - 91ms/epoch - 4ms/step\n",
      "Epoch 39/250\n",
      "24/24 - 0s - loss: 0.0201 - mse: 0.0201 - mae: 0.1055 - mape: 1509409.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 40/250\n",
      "24/24 - 0s - loss: 0.0175 - mse: 0.0175 - mae: 0.0962 - mape: 1117922.6250 - 92ms/epoch - 4ms/step\n",
      "Epoch 41/250\n",
      "24/24 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0926 - mape: 977065.3750 - 91ms/epoch - 4ms/step\n",
      "Epoch 42/250\n",
      "24/24 - 0s - loss: 0.0190 - mse: 0.0190 - mae: 0.0950 - mape: 959530.0625 - 91ms/epoch - 4ms/step\n",
      "Epoch 43/250\n",
      "24/24 - 0s - loss: 0.0177 - mse: 0.0177 - mae: 0.0935 - mape: 1254713.0000 - 92ms/epoch - 4ms/step\n",
      "Epoch 44/250\n",
      "24/24 - 0s - loss: 0.0149 - mse: 0.0149 - mae: 0.0898 - mape: 1282211.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 45/250\n",
      "24/24 - 0s - loss: 0.0182 - mse: 0.0182 - mae: 0.0952 - mape: 1131560.0000 - 92ms/epoch - 4ms/step\n",
      "Epoch 46/250\n",
      "24/24 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0888 - mape: 1186294.7500 - 92ms/epoch - 4ms/step\n",
      "Epoch 47/250\n",
      "24/24 - 0s - loss: 0.0169 - mse: 0.0169 - mae: 0.0933 - mape: 1015258.8125 - 92ms/epoch - 4ms/step\n",
      "Epoch 48/250\n",
      "24/24 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0900 - mape: 873230.8125 - 89ms/epoch - 4ms/step\n",
      "Epoch 49/250\n",
      "24/24 - 0s - loss: 0.0174 - mse: 0.0174 - mae: 0.0914 - mape: 1037554.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 50/250\n",
      "24/24 - 0s - loss: 0.0166 - mse: 0.0166 - mae: 0.1005 - mape: 823378.5625 - 92ms/epoch - 4ms/step\n",
      "Epoch 51/250\n",
      "24/24 - 0s - loss: 0.0156 - mse: 0.0156 - mae: 0.0858 - mape: 534497.3125 - 93ms/epoch - 4ms/step\n",
      "Epoch 52/250\n",
      "24/24 - 0s - loss: 0.0134 - mse: 0.0134 - mae: 0.0834 - mape: 1218616.6250 - 92ms/epoch - 4ms/step\n",
      "Epoch 53/250\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0911 - mape: 1270641.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 54/250\n",
      "24/24 - 0s - loss: 0.0147 - mse: 0.0147 - mae: 0.0901 - mape: 816528.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 55/250\n",
      "24/24 - 0s - loss: 0.0153 - mse: 0.0153 - mae: 0.0904 - mape: 761639.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 56/250\n",
      "24/24 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.0962 - mape: 1167063.3750 - 102ms/epoch - 4ms/step\n",
      "Epoch 57/250\n",
      "24/24 - 0s - loss: 0.0168 - mse: 0.0168 - mae: 0.1046 - mape: 657741.3750 - 83ms/epoch - 3ms/step\n",
      "Epoch 58/250\n",
      "24/24 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0887 - mape: 1484519.3750 - 91ms/epoch - 4ms/step\n",
      "Epoch 59/250\n",
      "24/24 - 0s - loss: 0.0138 - mse: 0.0138 - mae: 0.0863 - mape: 1106873.0000 - 99ms/epoch - 4ms/step\n",
      "Epoch 60/250\n",
      "24/24 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0904 - mape: 1357424.1250 - 102ms/epoch - 4ms/step\n",
      "Epoch 61/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0118 - mae: 0.0756 - mape: 921549.2500 - 109ms/epoch - 5ms/step\n",
      "Epoch 62/250\n",
      "24/24 - 0s - loss: 0.0146 - mse: 0.0146 - mae: 0.0844 - mape: 952971.8125 - 109ms/epoch - 5ms/step\n",
      "Epoch 63/250\n",
      "24/24 - 0s - loss: 0.0136 - mse: 0.0136 - mae: 0.0840 - mape: 1148828.3750 - 120ms/epoch - 5ms/step\n",
      "Epoch 64/250\n",
      "24/24 - 0s - loss: 0.0120 - mse: 0.0120 - mae: 0.0814 - mape: 727547.4375 - 121ms/epoch - 5ms/step\n",
      "Epoch 65/250\n",
      "24/24 - 0s - loss: 0.0130 - mse: 0.0130 - mae: 0.0849 - mape: 1422618.6250 - 119ms/epoch - 5ms/step\n",
      "Epoch 66/250\n",
      "24/24 - 0s - loss: 0.0157 - mse: 0.0157 - mae: 0.0929 - mape: 1469707.1250 - 121ms/epoch - 5ms/step\n",
      "Epoch 67/250\n",
      "24/24 - 0s - loss: 0.0125 - mse: 0.0125 - mae: 0.0801 - mape: 713857.4375 - 113ms/epoch - 5ms/step\n",
      "Epoch 68/250\n",
      "24/24 - 0s - loss: 0.0130 - mse: 0.0130 - mae: 0.0822 - mape: 594954.6875 - 121ms/epoch - 5ms/step\n",
      "Epoch 69/250\n",
      "24/24 - 0s - loss: 0.0126 - mse: 0.0126 - mae: 0.0821 - mape: 836715.8750 - 112ms/epoch - 5ms/step\n",
      "Epoch 70/250\n",
      "24/24 - 0s - loss: 0.0128 - mse: 0.0128 - mae: 0.0867 - mape: 596800.6875 - 112ms/epoch - 5ms/step\n",
      "Epoch 71/250\n",
      "24/24 - 0s - loss: 0.0125 - mse: 0.0125 - mae: 0.0827 - mape: 929953.6250 - 112ms/epoch - 5ms/step\n",
      "Epoch 72/250\n",
      "24/24 - 0s - loss: 0.0124 - mse: 0.0124 - mae: 0.0807 - mape: 1309177.1250 - 112ms/epoch - 5ms/step\n",
      "Epoch 73/250\n",
      "24/24 - 0s - loss: 0.0120 - mse: 0.0120 - mae: 0.0817 - mape: 1370456.2500 - 111ms/epoch - 5ms/step\n",
      "Epoch 74/250\n",
      "24/24 - 0s - loss: 0.0137 - mse: 0.0137 - mae: 0.0869 - mape: 897803.1250 - 112ms/epoch - 5ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 75/250\n",
      "24/24 - 0s - loss: 0.0095 - mse: 0.0095 - mae: 0.0743 - mape: 896045.1875 - 122ms/epoch - 5ms/step\n",
      "Epoch 76/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0118 - mae: 0.0802 - mape: 642344.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 77/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0118 - mae: 0.0799 - mape: 803678.8750 - 131ms/epoch - 5ms/step\n",
      "Epoch 78/250\n",
      "24/24 - 0s - loss: 0.0127 - mse: 0.0127 - mae: 0.0788 - mape: 755555.5000 - 122ms/epoch - 5ms/step\n",
      "Epoch 79/250\n",
      "24/24 - 0s - loss: 0.0097 - mse: 0.0097 - mae: 0.0742 - mape: 744463.5625 - 121ms/epoch - 5ms/step\n",
      "Epoch 80/250\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0122 - mae: 0.0801 - mape: 601576.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 81/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0755 - mape: 1075609.7500 - 119ms/epoch - 5ms/step\n",
      "Epoch 82/250\n",
      "24/24 - 0s - loss: 0.0110 - mse: 0.0110 - mae: 0.0785 - mape: 1158282.3750 - 127ms/epoch - 5ms/step\n",
      "Epoch 83/250\n",
      "24/24 - 0s - loss: 0.0115 - mse: 0.0115 - mae: 0.0803 - mape: 625449.1875 - 132ms/epoch - 6ms/step\n",
      "Epoch 84/250\n",
      "24/24 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0920 - mape: 861496.7500 - 131ms/epoch - 5ms/step\n",
      "Epoch 85/250\n",
      "24/24 - 0s - loss: 0.0108 - mse: 0.0108 - mae: 0.0773 - mape: 935802.1250 - 129ms/epoch - 5ms/step\n",
      "Epoch 86/250\n",
      "24/24 - 0s - loss: 0.0101 - mse: 0.0101 - mae: 0.0692 - mape: 1189745.8750 - 131ms/epoch - 5ms/step\n",
      "Epoch 87/250\n",
      "24/24 - 0s - loss: 0.0102 - mse: 0.0102 - mae: 0.0754 - mape: 1310425.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 88/250\n",
      "24/24 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0702 - mape: 1410307.5000 - 122ms/epoch - 5ms/step\n",
      "Epoch 89/250\n",
      "24/24 - 0s - loss: 0.0104 - mse: 0.0104 - mae: 0.0773 - mape: 1162450.1250 - 115ms/epoch - 5ms/step\n",
      "Epoch 90/250\n",
      "24/24 - 0s - loss: 0.0101 - mse: 0.0101 - mae: 0.0730 - mape: 1060132.7500 - 118ms/epoch - 5ms/step\n",
      "Epoch 91/250\n",
      "24/24 - 0s - loss: 0.0115 - mse: 0.0115 - mae: 0.0812 - mape: 920683.6875 - 122ms/epoch - 5ms/step\n",
      "Epoch 92/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0728 - mape: 1157348.6250 - 123ms/epoch - 5ms/step\n",
      "Epoch 93/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0783 - mape: 1039115.1875 - 124ms/epoch - 5ms/step\n",
      "Epoch 94/250\n",
      "24/24 - 0s - loss: 0.0086 - mse: 0.0086 - mae: 0.0734 - mape: 980233.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 95/250\n",
      "24/24 - 0s - loss: 0.0099 - mse: 0.0099 - mae: 0.0747 - mape: 1179159.8750 - 130ms/epoch - 5ms/step\n",
      "Epoch 96/250\n",
      "24/24 - 0s - loss: 0.0088 - mse: 0.0088 - mae: 0.0702 - mape: 729999.9375 - 121ms/epoch - 5ms/step\n",
      "Epoch 97/250\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0113 - mae: 0.0788 - mape: 655567.3750 - 118ms/epoch - 5ms/step\n",
      "Epoch 98/250\n",
      "24/24 - 0s - loss: 0.0097 - mse: 0.0097 - mae: 0.0750 - mape: 777686.9375 - 121ms/epoch - 5ms/step\n",
      "Epoch 99/250\n",
      "24/24 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0774 - mape: 682391.5625 - 140ms/epoch - 6ms/step\n",
      "Epoch 100/250\n",
      "24/24 - 0s - loss: 0.0088 - mse: 0.0088 - mae: 0.0725 - mape: 1035457.0625 - 122ms/epoch - 5ms/step\n",
      "Epoch 101/250\n",
      "24/24 - 0s - loss: 0.0095 - mse: 0.0095 - mae: 0.0753 - mape: 927362.5000 - 119ms/epoch - 5ms/step\n",
      "Epoch 102/250\n",
      "24/24 - 0s - loss: 0.0086 - mse: 0.0086 - mae: 0.0724 - mape: 756013.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 103/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0772 - mape: 1292861.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 104/250\n",
      "24/24 - 0s - loss: 0.0109 - mse: 0.0109 - mae: 0.0753 - mape: 972317.4375 - 122ms/epoch - 5ms/step\n",
      "Epoch 105/250\n",
      "24/24 - 0s - loss: 0.0097 - mse: 0.0097 - mae: 0.0736 - mape: 896148.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 106/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0692 - mape: 781568.1875 - 120ms/epoch - 5ms/step\n",
      "Epoch 107/250\n",
      "24/24 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0689 - mape: 912937.1250 - 120ms/epoch - 5ms/step\n",
      "Epoch 108/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0672 - mape: 802712.0000 - 123ms/epoch - 5ms/step\n",
      "Epoch 109/250\n",
      "24/24 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0714 - mape: 1472175.0000 - 121ms/epoch - 5ms/step\n",
      "Epoch 110/250\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0122 - mae: 0.0808 - mape: 756295.6250 - 119ms/epoch - 5ms/step\n",
      "Epoch 111/250\n",
      "24/24 - 0s - loss: 0.0092 - mse: 0.0092 - mae: 0.0754 - mape: 1107725.1250 - 129ms/epoch - 5ms/step\n",
      "Epoch 112/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0718 - mape: 1192839.0000 - 131ms/epoch - 5ms/step\n",
      "Epoch 113/250\n",
      "24/24 - 0s - loss: 0.0083 - mse: 0.0083 - mae: 0.0707 - mape: 814498.5000 - 115ms/epoch - 5ms/step\n",
      "Epoch 114/250\n",
      "24/24 - 0s - loss: 0.0102 - mse: 0.0102 - mae: 0.0757 - mape: 949646.3750 - 117ms/epoch - 5ms/step\n",
      "Epoch 115/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0722 - mape: 870730.2500 - 121ms/epoch - 5ms/step\n",
      "Epoch 116/250\n",
      "24/24 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0698 - mape: 910142.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 117/250\n",
      "24/24 - 0s - loss: 0.0078 - mse: 0.0078 - mae: 0.0652 - mape: 1085901.1250 - 120ms/epoch - 5ms/step\n",
      "Epoch 118/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0673 - mape: 689094.2500 - 121ms/epoch - 5ms/step\n",
      "Epoch 119/250\n",
      "24/24 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0657 - mape: 1013438.9375 - 129ms/epoch - 5ms/step\n",
      "Epoch 120/250\n",
      "24/24 - 0s - loss: 0.0106 - mse: 0.0106 - mae: 0.0774 - mape: 694634.2500 - 121ms/epoch - 5ms/step\n",
      "Epoch 121/250\n",
      "24/24 - 0s - loss: 0.0086 - mse: 0.0086 - mae: 0.0692 - mape: 824105.7500 - 119ms/epoch - 5ms/step\n",
      "Epoch 122/250\n",
      "24/24 - 0s - loss: 0.0130 - mse: 0.0130 - mae: 0.0781 - mape: 977953.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 123/250\n",
      "24/24 - 0s - loss: 0.0084 - mse: 0.0084 - mae: 0.0667 - mape: 1058746.8750 - 122ms/epoch - 5ms/step\n",
      "Epoch 124/250\n",
      "24/24 - 0s - loss: 0.0099 - mse: 0.0099 - mae: 0.0719 - mape: 722716.1875 - 129ms/epoch - 5ms/step\n",
      "Epoch 125/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0645 - mape: 1295251.6250 - 125ms/epoch - 5ms/step\n",
      "Epoch 126/250\n",
      "24/24 - 0s - loss: 0.0092 - mse: 0.0092 - mae: 0.0724 - mape: 1227969.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 127/250\n",
      "24/24 - 0s - loss: 0.0096 - mse: 0.0096 - mae: 0.0729 - mape: 1123135.1250 - 124ms/epoch - 5ms/step\n",
      "Epoch 128/250\n",
      "24/24 - 0s - loss: 0.0118 - mse: 0.0118 - mae: 0.0780 - mape: 1065924.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 129/250\n",
      "24/24 - 0s - loss: 0.0111 - mse: 0.0111 - mae: 0.0784 - mape: 964636.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 130/250\n",
      "24/24 - 0s - loss: 0.0084 - mse: 0.0084 - mae: 0.0692 - mape: 871630.8750 - 124ms/epoch - 5ms/step\n",
      "Epoch 131/250\n",
      "24/24 - 0s - loss: 0.0095 - mse: 0.0095 - mae: 0.0665 - mape: 1292668.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 132/250\n",
      "24/24 - 0s - loss: 0.0087 - mse: 0.0087 - mae: 0.0657 - mape: 1225039.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 133/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0620 - mape: 703651.8125 - 122ms/epoch - 5ms/step\n",
      "Epoch 134/250\n",
      "24/24 - 0s - loss: 0.0097 - mse: 0.0097 - mae: 0.0721 - mape: 929541.5625 - 124ms/epoch - 5ms/step\n",
      "Epoch 135/250\n",
      "24/24 - 0s - loss: 0.0099 - mse: 0.0099 - mae: 0.0771 - mape: 1030077.6875 - 136ms/epoch - 6ms/step\n",
      "Epoch 136/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0725 - mape: 989811.1875 - 124ms/epoch - 5ms/step\n",
      "Epoch 137/250\n",
      "24/24 - 0s - loss: 0.0094 - mse: 0.0094 - mae: 0.0722 - mape: 787877.8750 - 121ms/epoch - 5ms/step\n",
      "Epoch 138/250\n",
      "24/24 - 0s - loss: 0.0074 - mse: 0.0074 - mae: 0.0665 - mape: 718483.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 139/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0602 - mape: 708050.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 140/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0620 - mape: 874246.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 141/250\n",
      "24/24 - 0s - loss: 0.0074 - mse: 0.0074 - mae: 0.0657 - mape: 967905.3750 - 129ms/epoch - 5ms/step\n",
      "Epoch 142/250\n",
      "24/24 - 0s - loss: 0.0093 - mse: 0.0093 - mae: 0.0676 - mape: 1276915.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 143/250\n",
      "24/24 - 0s - loss: 0.0085 - mse: 0.0085 - mae: 0.0688 - mape: 638810.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 144/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0623 - mape: 557549.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 145/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0657 - mape: 815468.8125 - 121ms/epoch - 5ms/step\n",
      "Epoch 146/250\n",
      "24/24 - 0s - loss: 0.0103 - mse: 0.0103 - mae: 0.0708 - mape: 1219971.2500 - 119ms/epoch - 5ms/step\n",
      "Epoch 147/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0663 - mape: 891706.5000 - 123ms/epoch - 5ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 148/250\n",
      "24/24 - 0s - loss: 0.0090 - mse: 0.0090 - mae: 0.0728 - mape: 893567.0625 - 122ms/epoch - 5ms/step\n",
      "Epoch 149/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0637 - mape: 1176218.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 150/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0622 - mape: 1080129.3750 - 126ms/epoch - 5ms/step\n",
      "Epoch 151/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0581 - mape: 650683.0000 - 127ms/epoch - 5ms/step\n",
      "Epoch 152/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0624 - mape: 752506.1250 - 124ms/epoch - 5ms/step\n",
      "Epoch 153/250\n",
      "24/24 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0627 - mape: 663870.5625 - 128ms/epoch - 5ms/step\n",
      "Epoch 154/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0587 - mape: 947056.0000 - 126ms/epoch - 5ms/step\n",
      "Epoch 155/250\n",
      "24/24 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0627 - mape: 1371762.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 156/250\n",
      "24/24 - 0s - loss: 0.0089 - mse: 0.0089 - mae: 0.0697 - mape: 584935.1875 - 123ms/epoch - 5ms/step\n",
      "Epoch 157/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0625 - mape: 646212.3750 - 123ms/epoch - 5ms/step\n",
      "Epoch 158/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0619 - mape: 991801.3125 - 121ms/epoch - 5ms/step\n",
      "Epoch 159/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0598 - mape: 699078.0000 - 125ms/epoch - 5ms/step\n",
      "Epoch 160/250\n",
      "24/24 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0655 - mape: 780238.9375 - 130ms/epoch - 5ms/step\n",
      "Epoch 161/250\n",
      "24/24 - 0s - loss: 0.0082 - mse: 0.0082 - mae: 0.0658 - mape: 627570.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 162/250\n",
      "24/24 - 0s - loss: 0.0078 - mse: 0.0078 - mae: 0.0613 - mape: 1198298.2500 - 127ms/epoch - 5ms/step\n",
      "Epoch 163/250\n",
      "24/24 - 0s - loss: 0.0084 - mse: 0.0084 - mae: 0.0677 - mape: 1378098.6250 - 124ms/epoch - 5ms/step\n",
      "Epoch 164/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0667 - mape: 704314.0625 - 121ms/epoch - 5ms/step\n",
      "Epoch 165/250\n",
      "24/24 - 0s - loss: 0.0079 - mse: 0.0079 - mae: 0.0619 - mape: 844456.1875 - 122ms/epoch - 5ms/step\n",
      "Epoch 166/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0628 - mape: 932078.6875 - 130ms/epoch - 5ms/step\n",
      "Epoch 167/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0590 - mape: 1017203.4375 - 121ms/epoch - 5ms/step\n",
      "Epoch 168/250\n",
      "24/24 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0652 - mape: 810065.7500 - 117ms/epoch - 5ms/step\n",
      "Epoch 169/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0583 - mape: 1363579.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 170/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0552 - mape: 1222323.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 171/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0675 - mape: 688723.2500 - 120ms/epoch - 5ms/step\n",
      "Epoch 172/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0620 - mape: 1473657.5000 - 122ms/epoch - 5ms/step\n",
      "Epoch 173/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0640 - mape: 919014.4375 - 118ms/epoch - 5ms/step\n",
      "Epoch 174/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0553 - mape: 882982.0000 - 121ms/epoch - 5ms/step\n",
      "Epoch 175/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0646 - mape: 1105701.1250 - 125ms/epoch - 5ms/step\n",
      "Epoch 176/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0608 - mape: 632634.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 177/250\n",
      "24/24 - 0s - loss: 0.0056 - mse: 0.0056 - mae: 0.0593 - mape: 738078.8750 - 124ms/epoch - 5ms/step\n",
      "Epoch 178/250\n",
      "24/24 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0610 - mape: 1414149.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 179/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0590 - mape: 1156225.5000 - 130ms/epoch - 5ms/step\n",
      "Epoch 180/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0581 - mape: 877081.1250 - 131ms/epoch - 5ms/step\n",
      "Epoch 181/250\n",
      "24/24 - 0s - loss: 0.0069 - mse: 0.0069 - mae: 0.0602 - mape: 1018966.2500 - 142ms/epoch - 6ms/step\n",
      "Epoch 182/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0644 - mape: 1265413.5000 - 124ms/epoch - 5ms/step\n",
      "Epoch 183/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0524 - mape: 922862.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 184/250\n",
      "24/24 - 0s - loss: 0.0145 - mse: 0.0145 - mae: 0.0729 - mape: 700385.7500 - 123ms/epoch - 5ms/step\n",
      "Epoch 185/250\n",
      "24/24 - 0s - loss: 0.0107 - mse: 0.0107 - mae: 0.0759 - mape: 1091868.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 186/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0599 - mape: 1264475.0000 - 119ms/epoch - 5ms/step\n",
      "Epoch 187/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0583 - mape: 630294.8125 - 118ms/epoch - 5ms/step\n",
      "Epoch 188/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0632 - mape: 863513.5625 - 122ms/epoch - 5ms/step\n",
      "Epoch 189/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0594 - mape: 804909.0625 - 129ms/epoch - 5ms/step\n",
      "Epoch 190/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0576 - mape: 659785.6875 - 124ms/epoch - 5ms/step\n",
      "Epoch 191/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0551 - mape: 884840.5000 - 126ms/epoch - 5ms/step\n",
      "Epoch 192/250\n",
      "24/24 - 0s - loss: 0.0049 - mse: 0.0049 - mae: 0.0531 - mape: 924611.4375 - 125ms/epoch - 5ms/step\n",
      "Epoch 193/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0583 - mape: 877799.0625 - 123ms/epoch - 5ms/step\n",
      "Epoch 194/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0592 - mape: 844532.5625 - 124ms/epoch - 5ms/step\n",
      "Epoch 195/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0545 - mape: 909947.1875 - 122ms/epoch - 5ms/step\n",
      "Epoch 196/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0546 - mape: 612605.1875 - 122ms/epoch - 5ms/step\n",
      "Epoch 197/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0579 - mape: 755131.2500 - 131ms/epoch - 5ms/step\n",
      "Epoch 198/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0565 - mape: 781882.5625 - 132ms/epoch - 6ms/step\n",
      "Epoch 199/250\n",
      "24/24 - 0s - loss: 0.0052 - mse: 0.0052 - mae: 0.0575 - mape: 879325.6875 - 133ms/epoch - 6ms/step\n",
      "Epoch 200/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0576 - mape: 676784.5000 - 130ms/epoch - 5ms/step\n",
      "Epoch 201/250\n",
      "24/24 - 0s - loss: 0.0062 - mse: 0.0062 - mae: 0.0582 - mape: 1207421.5000 - 132ms/epoch - 6ms/step\n",
      "Epoch 202/250\n",
      "24/24 - 0s - loss: 0.0054 - mse: 0.0054 - mae: 0.0540 - mape: 795791.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 203/250\n",
      "24/24 - 0s - loss: 0.0067 - mse: 0.0067 - mae: 0.0572 - mape: 709189.7500 - 127ms/epoch - 5ms/step\n",
      "Epoch 204/250\n",
      "24/24 - 0s - loss: 0.0068 - mse: 0.0068 - mae: 0.0558 - mape: 1190366.3750 - 124ms/epoch - 5ms/step\n",
      "Epoch 205/250\n",
      "24/24 - 0s - loss: 0.0071 - mse: 0.0071 - mae: 0.0596 - mape: 585471.3750 - 133ms/epoch - 6ms/step\n",
      "Epoch 206/250\n",
      "24/24 - 0s - loss: 0.0076 - mse: 0.0076 - mae: 0.0635 - mape: 1007124.8750 - 130ms/epoch - 5ms/step\n",
      "Epoch 207/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0580 - mape: 991250.4375 - 138ms/epoch - 6ms/step\n",
      "Epoch 208/250\n",
      "24/24 - 0s - loss: 0.0047 - mse: 0.0047 - mae: 0.0523 - mape: 820133.4375 - 171ms/epoch - 7ms/step\n",
      "Epoch 209/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0591 - mape: 560423.1250 - 132ms/epoch - 5ms/step\n",
      "Epoch 210/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0608 - mape: 601064.6250 - 117ms/epoch - 5ms/step\n",
      "Epoch 211/250\n",
      "24/24 - 0s - loss: 0.0063 - mse: 0.0063 - mae: 0.0603 - mape: 872420.0625 - 122ms/epoch - 5ms/step\n",
      "Epoch 212/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0614 - mape: 686581.0000 - 123ms/epoch - 5ms/step\n",
      "Epoch 213/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0598 - mape: 792734.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 214/250\n",
      "24/24 - 0s - loss: 0.0072 - mse: 0.0072 - mae: 0.0569 - mape: 805291.6250 - 135ms/epoch - 6ms/step\n",
      "Epoch 215/250\n",
      "24/24 - 0s - loss: 0.0061 - mse: 0.0061 - mae: 0.0600 - mape: 1015869.3125 - 120ms/epoch - 5ms/step\n",
      "Epoch 216/250\n",
      "24/24 - 0s - loss: 0.0073 - mse: 0.0073 - mae: 0.0578 - mape: 848150.9375 - 121ms/epoch - 5ms/step\n",
      "Epoch 217/250\n",
      "24/24 - 0s - loss: 0.0058 - mse: 0.0058 - mae: 0.0585 - mape: 1325486.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 218/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0576 - mape: 998362.0625 - 122ms/epoch - 5ms/step\n",
      "Epoch 219/250\n",
      "24/24 - 0s - loss: 0.0084 - mse: 0.0084 - mae: 0.0586 - mape: 759569.8125 - 129ms/epoch - 5ms/step\n",
      "Epoch 220/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0662 - mape: 1080538.3750 - 123ms/epoch - 5ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 221/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0570 - mape: 894444.3125 - 122ms/epoch - 5ms/step\n",
      "Epoch 222/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0529 - mape: 1164417.8750 - 118ms/epoch - 5ms/step\n",
      "Epoch 223/250\n",
      "24/24 - 0s - loss: 0.0077 - mse: 0.0077 - mae: 0.0579 - mape: 1071884.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 224/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0620 - mape: 855984.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 225/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0577 - mape: 1431976.1250 - 124ms/epoch - 5ms/step\n",
      "Epoch 226/250\n",
      "24/24 - 0s - loss: 0.0044 - mse: 0.0044 - mae: 0.0500 - mape: 641840.3125 - 130ms/epoch - 5ms/step\n",
      "Epoch 227/250\n",
      "24/24 - 0s - loss: 0.0045 - mse: 0.0045 - mae: 0.0503 - mape: 655732.2500 - 121ms/epoch - 5ms/step\n",
      "Epoch 228/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0593 - mape: 665824.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 229/250\n",
      "24/24 - 0s - loss: 0.0042 - mse: 0.0042 - mae: 0.0473 - mape: 844491.1875 - 130ms/epoch - 5ms/step\n",
      "Epoch 230/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0521 - mape: 729268.0625 - 123ms/epoch - 5ms/step\n",
      "Epoch 231/250\n",
      "24/24 - 0s - loss: 0.0043 - mse: 0.0043 - mae: 0.0508 - mape: 888497.7500 - 131ms/epoch - 5ms/step\n",
      "Epoch 232/250\n",
      "24/24 - 0s - loss: 0.0050 - mse: 0.0050 - mae: 0.0526 - mape: 566943.0625 - 123ms/epoch - 5ms/step\n",
      "Epoch 233/250\n",
      "24/24 - 0s - loss: 0.0048 - mse: 0.0048 - mae: 0.0525 - mape: 584639.8125 - 131ms/epoch - 5ms/step\n",
      "Epoch 234/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0533 - mape: 1366751.0000 - 124ms/epoch - 5ms/step\n",
      "Epoch 235/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0565 - mape: 899512.6875 - 124ms/epoch - 5ms/step\n",
      "Epoch 236/250\n",
      "24/24 - 0s - loss: 0.0075 - mse: 0.0075 - mae: 0.0637 - mape: 673396.1875 - 124ms/epoch - 5ms/step\n",
      "Epoch 237/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0511 - mape: 845633.1875 - 124ms/epoch - 5ms/step\n",
      "Epoch 238/250\n",
      "24/24 - 0s - loss: 0.0065 - mse: 0.0065 - mae: 0.0615 - mape: 589191.3125 - 131ms/epoch - 5ms/step\n",
      "Epoch 239/250\n",
      "24/24 - 0s - loss: 0.0046 - mse: 0.0046 - mae: 0.0500 - mape: 725354.9375 - 131ms/epoch - 5ms/step\n",
      "Epoch 240/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0516 - mape: 1164501.6250 - 131ms/epoch - 5ms/step\n",
      "Epoch 241/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0567 - mape: 1075979.3750 - 124ms/epoch - 5ms/step\n",
      "Epoch 242/250\n",
      "24/24 - 0s - loss: 0.0064 - mse: 0.0064 - mae: 0.0562 - mape: 747800.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 243/250\n",
      "24/24 - 0s - loss: 0.0055 - mse: 0.0055 - mae: 0.0534 - mape: 540086.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 244/250\n",
      "24/24 - 0s - loss: 0.0057 - mse: 0.0057 - mae: 0.0549 - mape: 687269.5625 - 131ms/epoch - 5ms/step\n",
      "Epoch 245/250\n",
      "24/24 - 0s - loss: 0.0080 - mse: 0.0080 - mae: 0.0600 - mape: 1428774.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 246/250\n",
      "24/24 - 0s - loss: 0.0070 - mse: 0.0070 - mae: 0.0612 - mape: 938075.1875 - 121ms/epoch - 5ms/step\n",
      "Epoch 247/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0560 - mape: 718348.5000 - 122ms/epoch - 5ms/step\n",
      "Epoch 248/250\n",
      "24/24 - 0s - loss: 0.0051 - mse: 0.0051 - mae: 0.0513 - mape: 567048.6250 - 121ms/epoch - 5ms/step\n",
      "Epoch 249/250\n",
      "24/24 - 0s - loss: 0.0059 - mse: 0.0059 - mae: 0.0573 - mape: 567569.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 250/250\n",
      "24/24 - 0s - loss: 0.0060 - mse: 0.0060 - mae: 0.0584 - mape: 1875988.0000 - 122ms/epoch - 5ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2954c9256c0>"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 4 include dropouts for overfitting - better\n",
    "from tensorflow.keras.layers import Dropout\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, activation='relu', input_shape=(1, look_back), return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(50, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mse', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=250, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "f5f353ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "24/24 - 5s - loss: 0.0688 - mse: 0.0688 - mae: 0.1931 - mape: 522692.0938 - 5s/epoch - 219ms/step\n",
      "Epoch 2/100\n",
      "24/24 - 0s - loss: 0.0268 - mse: 0.0268 - mae: 0.1131 - mape: 2090471.6250 - 111ms/epoch - 5ms/step\n",
      "Epoch 3/100\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1165 - mape: 1916004.8750 - 112ms/epoch - 5ms/step\n",
      "Epoch 4/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1112 - mape: 1655890.7500 - 111ms/epoch - 5ms/step\n",
      "Epoch 5/100\n",
      "24/24 - 0s - loss: 0.0251 - mse: 0.0251 - mae: 0.1097 - mape: 1895888.0000 - 115ms/epoch - 5ms/step\n",
      "Epoch 6/100\n",
      "24/24 - 0s - loss: 0.0248 - mse: 0.0248 - mae: 0.1105 - mape: 1584779.0000 - 121ms/epoch - 5ms/step\n",
      "Epoch 7/100\n",
      "24/24 - 0s - loss: 0.0269 - mse: 0.0269 - mae: 0.1174 - mape: 2056144.8750 - 111ms/epoch - 5ms/step\n",
      "Epoch 8/100\n",
      "24/24 - 0s - loss: 0.0240 - mse: 0.0240 - mae: 0.1116 - mape: 1962348.8750 - 112ms/epoch - 5ms/step\n",
      "Epoch 9/100\n",
      "24/24 - 0s - loss: 0.0262 - mse: 0.0262 - mae: 0.1105 - mape: 1309016.3750 - 119ms/epoch - 5ms/step\n",
      "Epoch 10/100\n",
      "24/24 - 0s - loss: 0.0222 - mse: 0.0222 - mae: 0.1118 - mape: 1962678.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 11/100\n",
      "24/24 - 0s - loss: 0.0240 - mse: 0.0240 - mae: 0.1122 - mape: 1699549.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 12/100\n",
      "24/24 - 0s - loss: 0.0228 - mse: 0.0228 - mae: 0.1053 - mape: 1817658.5000 - 123ms/epoch - 5ms/step\n",
      "Epoch 13/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1079 - mape: 1713126.2500 - 123ms/epoch - 5ms/step\n",
      "Epoch 14/100\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1111 - mape: 1564292.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 15/100\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1031 - mape: 1484940.2500 - 123ms/epoch - 5ms/step\n",
      "Epoch 16/100\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1185 - mape: 1756605.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 17/100\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1075 - mape: 1435592.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 18/100\n",
      "24/24 - 0s - loss: 0.0243 - mse: 0.0243 - mae: 0.1119 - mape: 1753370.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 19/100\n",
      "24/24 - 0s - loss: 0.0251 - mse: 0.0251 - mae: 0.1098 - mape: 1194532.3750 - 123ms/epoch - 5ms/step\n",
      "Epoch 20/100\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1050 - mape: 1616241.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 21/100\n",
      "24/24 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1100 - mape: 1489675.3750 - 115ms/epoch - 5ms/step\n",
      "Epoch 22/100\n",
      "24/24 - 0s - loss: 0.0233 - mse: 0.0233 - mae: 0.1036 - mape: 1360602.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 23/100\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1174 - mape: 1727552.0000 - 123ms/epoch - 5ms/step\n",
      "Epoch 24/100\n",
      "24/24 - 0s - loss: 0.0245 - mse: 0.0245 - mae: 0.1075 - mape: 1233795.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 25/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1115 - mape: 1645148.3750 - 131ms/epoch - 5ms/step\n",
      "Epoch 26/100\n",
      "24/24 - 0s - loss: 0.0248 - mse: 0.0248 - mae: 0.1163 - mape: 1646867.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 27/100\n",
      "24/24 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1079 - mape: 1585611.0000 - 114ms/epoch - 5ms/step\n",
      "Epoch 28/100\n",
      "24/24 - 0s - loss: 0.0228 - mse: 0.0228 - mae: 0.1093 - mape: 1374099.0000 - 114ms/epoch - 5ms/step\n",
      "Epoch 29/100\n",
      "24/24 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1131 - mape: 1485410.2500 - 119ms/epoch - 5ms/step\n",
      "Epoch 30/100\n",
      "24/24 - 0s - loss: 0.0243 - mse: 0.0243 - mae: 0.1086 - mape: 1279230.1250 - 112ms/epoch - 5ms/step\n",
      "Epoch 31/100\n",
      "24/24 - 0s - loss: 0.0240 - mse: 0.0240 - mae: 0.1058 - mape: 1253925.8750 - 101ms/epoch - 4ms/step\n",
      "Epoch 32/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1012 - mape: 1372251.7500 - 102ms/epoch - 4ms/step\n",
      "Epoch 33/100\n",
      "24/24 - 0s - loss: 0.0246 - mse: 0.0246 - mae: 0.1081 - mape: 1589139.1250 - 109ms/epoch - 5ms/step\n",
      "Epoch 34/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1098 - mape: 1230489.2500 - 111ms/epoch - 5ms/step\n",
      "Epoch 35/100\n",
      "24/24 - 0s - loss: 0.0248 - mse: 0.0248 - mae: 0.1089 - mape: 1228454.6250 - 112ms/epoch - 5ms/step\n",
      "Epoch 36/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1066 - mape: 1383489.8750 - 130ms/epoch - 5ms/step\n",
      "Epoch 37/100\n",
      "24/24 - 0s - loss: 0.0232 - mse: 0.0232 - mae: 0.1129 - mape: 1558574.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 38/100\n",
      "24/24 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1043 - mape: 1323367.6250 - 132ms/epoch - 5ms/step\n",
      "Epoch 39/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1103 - mape: 1578300.5000 - 143ms/epoch - 6ms/step\n",
      "Epoch 40/100\n",
      "24/24 - 0s - loss: 0.0222 - mse: 0.0222 - mae: 0.1076 - mape: 1666235.8750 - 133ms/epoch - 6ms/step\n",
      "Epoch 41/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1049 - mape: 1307924.8750 - 140ms/epoch - 6ms/step\n",
      "Epoch 42/100\n",
      "24/24 - 0s - loss: 0.0243 - mse: 0.0243 - mae: 0.1080 - mape: 1421534.8750 - 137ms/epoch - 6ms/step\n",
      "Epoch 43/100\n",
      "24/24 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1071 - mape: 1387845.5000 - 143ms/epoch - 6ms/step\n",
      "Epoch 44/100\n",
      "24/24 - 0s - loss: 0.0221 - mse: 0.0221 - mae: 0.1080 - mape: 1468506.8750 - 142ms/epoch - 6ms/step\n",
      "Epoch 45/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1021 - mape: 1496605.6250 - 142ms/epoch - 6ms/step\n",
      "Epoch 46/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1046 - mape: 1459986.3750 - 144ms/epoch - 6ms/step\n",
      "Epoch 47/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1082 - mape: 1541861.5000 - 144ms/epoch - 6ms/step\n",
      "Epoch 48/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1059 - mape: 1315232.3750 - 143ms/epoch - 6ms/step\n",
      "Epoch 49/100\n",
      "24/24 - 0s - loss: 0.0238 - mse: 0.0238 - mae: 0.1033 - mape: 1166534.0000 - 143ms/epoch - 6ms/step\n",
      "Epoch 50/100\n",
      "24/24 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1125 - mape: 1346803.7500 - 142ms/epoch - 6ms/step\n",
      "Epoch 51/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1028 - mape: 1263556.1250 - 133ms/epoch - 6ms/step\n",
      "Epoch 52/100\n",
      "24/24 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1053 - mape: 1623047.3750 - 141ms/epoch - 6ms/step\n",
      "Epoch 53/100\n",
      "24/24 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1062 - mape: 1289598.5000 - 142ms/epoch - 6ms/step\n",
      "Epoch 54/100\n",
      "24/24 - 0s - loss: 0.0227 - mse: 0.0227 - mae: 0.1055 - mape: 1282914.7500 - 140ms/epoch - 6ms/step\n",
      "Epoch 55/100\n",
      "24/24 - 0s - loss: 0.0237 - mse: 0.0237 - mae: 0.1115 - mape: 1219943.8750 - 143ms/epoch - 6ms/step\n",
      "Epoch 56/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1097 - mape: 1534620.0000 - 135ms/epoch - 6ms/step\n",
      "Epoch 57/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1061 - mape: 1332150.3750 - 138ms/epoch - 6ms/step\n",
      "Epoch 58/100\n",
      "24/24 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1035 - mape: 1631817.6250 - 134ms/epoch - 6ms/step\n",
      "Epoch 59/100\n",
      "24/24 - 0s - loss: 0.0232 - mse: 0.0232 - mae: 0.1110 - mape: 1455371.0000 - 132ms/epoch - 6ms/step\n",
      "Epoch 60/100\n",
      "24/24 - 0s - loss: 0.0227 - mse: 0.0227 - mae: 0.1051 - mape: 1379472.5000 - 131ms/epoch - 5ms/step\n",
      "Epoch 61/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1072 - mape: 1216963.0000 - 140ms/epoch - 6ms/step\n",
      "Epoch 62/100\n",
      "24/24 - 0s - loss: 0.0241 - mse: 0.0241 - mae: 0.1070 - mape: 1398709.1250 - 142ms/epoch - 6ms/step\n",
      "Epoch 63/100\n",
      "24/24 - 0s - loss: 0.0237 - mse: 0.0237 - mae: 0.1142 - mape: 1677315.2500 - 142ms/epoch - 6ms/step\n",
      "Epoch 64/100\n",
      "24/24 - 0s - loss: 0.0211 - mse: 0.0211 - mae: 0.1046 - mape: 1504157.7500 - 141ms/epoch - 6ms/step\n",
      "Epoch 65/100\n",
      "24/24 - 0s - loss: 0.0227 - mse: 0.0227 - mae: 0.1115 - mape: 1528812.5000 - 141ms/epoch - 6ms/step\n",
      "Epoch 66/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1107 - mape: 1567289.2500 - 140ms/epoch - 6ms/step\n",
      "Epoch 67/100\n",
      "24/24 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1032 - mape: 1249184.2500 - 142ms/epoch - 6ms/step\n",
      "Epoch 68/100\n",
      "24/24 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1065 - mape: 1376715.8750 - 152ms/epoch - 6ms/step\n",
      "Epoch 69/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1066 - mape: 1355975.6250 - 133ms/epoch - 6ms/step\n",
      "Epoch 70/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1029 - mape: 1194983.2500 - 143ms/epoch - 6ms/step\n",
      "Epoch 71/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1063 - mape: 1406421.8750 - 141ms/epoch - 6ms/step\n",
      "Epoch 72/100\n",
      "24/24 - 0s - loss: 0.0232 - mse: 0.0232 - mae: 0.1052 - mape: 1395716.0000 - 143ms/epoch - 6ms/step\n",
      "Epoch 73/100\n",
      "24/24 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.1002 - mape: 1378640.7500 - 140ms/epoch - 6ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "24/24 - 0s - loss: 0.0243 - mse: 0.0243 - mae: 0.1119 - mape: 1511542.2500 - 142ms/epoch - 6ms/step\n",
      "Epoch 75/100\n",
      "24/24 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1023 - mape: 1380695.3750 - 143ms/epoch - 6ms/step\n",
      "Epoch 76/100\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1048 - mape: 1382565.1250 - 152ms/epoch - 6ms/step\n",
      "Epoch 77/100\n",
      "24/24 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.1035 - mape: 1307799.2500 - 151ms/epoch - 6ms/step\n",
      "Epoch 78/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1062 - mape: 1116940.1250 - 150ms/epoch - 6ms/step\n",
      "Epoch 79/100\n",
      "24/24 - 0s - loss: 0.0200 - mse: 0.0200 - mae: 0.1011 - mape: 1401460.0000 - 154ms/epoch - 6ms/step\n",
      "Epoch 80/100\n",
      "24/24 - 0s - loss: 0.0233 - mse: 0.0233 - mae: 0.1069 - mape: 1254410.8750 - 150ms/epoch - 6ms/step\n",
      "Epoch 81/100\n",
      "24/24 - 0s - loss: 0.0227 - mse: 0.0227 - mae: 0.1054 - mape: 1074689.8750 - 151ms/epoch - 6ms/step\n",
      "Epoch 82/100\n",
      "24/24 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1011 - mape: 1125009.6250 - 151ms/epoch - 6ms/step\n",
      "Epoch 83/100\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1114 - mape: 1400746.6250 - 152ms/epoch - 6ms/step\n",
      "Epoch 84/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1048 - mape: 1280538.6250 - 150ms/epoch - 6ms/step\n",
      "Epoch 85/100\n",
      "24/24 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.1019 - mape: 1087104.2500 - 145ms/epoch - 6ms/step\n",
      "Epoch 86/100\n",
      "24/24 - 0s - loss: 0.0213 - mse: 0.0213 - mae: 0.1052 - mape: 1495691.8750 - 142ms/epoch - 6ms/step\n",
      "Epoch 87/100\n",
      "24/24 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1026 - mape: 1299691.5000 - 142ms/epoch - 6ms/step\n",
      "Epoch 88/100\n",
      "24/24 - 0s - loss: 0.0211 - mse: 0.0211 - mae: 0.1042 - mape: 1448894.5000 - 151ms/epoch - 6ms/step\n",
      "Epoch 89/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1033 - mape: 1368090.3750 - 148ms/epoch - 6ms/step\n",
      "Epoch 90/100\n",
      "24/24 - 0s - loss: 0.0215 - mse: 0.0215 - mae: 0.1045 - mape: 1222766.3750 - 146ms/epoch - 6ms/step\n",
      "Epoch 91/100\n",
      "24/24 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1043 - mape: 1118271.8750 - 143ms/epoch - 6ms/step\n",
      "Epoch 92/100\n",
      "24/24 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1068 - mape: 1266150.7500 - 142ms/epoch - 6ms/step\n",
      "Epoch 93/100\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1065 - mape: 1158767.0000 - 141ms/epoch - 6ms/step\n",
      "Epoch 94/100\n",
      "24/24 - 0s - loss: 0.0215 - mse: 0.0215 - mae: 0.1080 - mape: 1443702.0000 - 143ms/epoch - 6ms/step\n",
      "Epoch 95/100\n",
      "24/24 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1054 - mape: 1388960.8750 - 151ms/epoch - 6ms/step\n",
      "Epoch 96/100\n",
      "24/24 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1054 - mape: 1272104.2500 - 146ms/epoch - 6ms/step\n",
      "Epoch 97/100\n",
      "24/24 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1022 - mape: 968032.7500 - 144ms/epoch - 6ms/step\n",
      "Epoch 98/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1049 - mape: 1444990.2500 - 141ms/epoch - 6ms/step\n",
      "Epoch 99/100\n",
      "24/24 - 0s - loss: 0.0224 - mse: 0.0224 - mae: 0.1061 - mape: 977833.3125 - 141ms/epoch - 6ms/step\n",
      "Epoch 100/100\n",
      "24/24 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1074 - mape: 1206524.2500 - 144ms/epoch - 6ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x295609e3e50>"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 5 tanh activation - worse\n",
    "from tensorflow.keras.layers import Dropout\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, activation='tanh', input_shape=(1, look_back), return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(50, activation='tanh'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mse', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "108a8f4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "24/24 - 3s - loss: 0.0387 - mse: 0.0387 - mae: 0.1536 - mape: 2591397.2500 - 3s/epoch - 131ms/step\n",
      "Epoch 2/100\n",
      "24/24 - 0s - loss: 0.0333 - mse: 0.0333 - mae: 0.1346 - mape: 2646226.5000 - 100ms/epoch - 4ms/step\n",
      "Epoch 3/100\n",
      "24/24 - 0s - loss: 0.0302 - mse: 0.0302 - mae: 0.1231 - mape: 2323012.7500 - 102ms/epoch - 4ms/step\n",
      "Epoch 4/100\n",
      "24/24 - 0s - loss: 0.0246 - mse: 0.0246 - mae: 0.1150 - mape: 2161660.0000 - 101ms/epoch - 4ms/step\n",
      "Epoch 5/100\n",
      "24/24 - 0s - loss: 0.0291 - mse: 0.0291 - mae: 0.1239 - mape: 2098232.0000 - 99ms/epoch - 4ms/step\n",
      "Epoch 6/100\n",
      "24/24 - 0s - loss: 0.0284 - mse: 0.0284 - mae: 0.1168 - mape: 1898482.1250 - 101ms/epoch - 4ms/step\n",
      "Epoch 7/100\n",
      "24/24 - 0s - loss: 0.0283 - mse: 0.0283 - mae: 0.1206 - mape: 1747844.8750 - 102ms/epoch - 4ms/step\n",
      "Epoch 8/100\n",
      "24/24 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1174 - mape: 2322115.5000 - 94ms/epoch - 4ms/step\n",
      "Epoch 9/100\n",
      "24/24 - 0s - loss: 0.0263 - mse: 0.0263 - mae: 0.1148 - mape: 1056723.6250 - 94ms/epoch - 4ms/step\n",
      "Epoch 10/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1208 - mape: 1886590.6250 - 94ms/epoch - 4ms/step\n",
      "Epoch 11/100\n",
      "24/24 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1120 - mape: 1495738.8750 - 93ms/epoch - 4ms/step\n",
      "Epoch 12/100\n",
      "24/24 - 0s - loss: 0.0237 - mse: 0.0237 - mae: 0.1090 - mape: 1626054.5000 - 99ms/epoch - 4ms/step\n",
      "Epoch 13/100\n",
      "24/24 - 0s - loss: 0.0253 - mse: 0.0253 - mae: 0.1108 - mape: 2000422.7500 - 102ms/epoch - 4ms/step\n",
      "Epoch 14/100\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1080 - mape: 1232963.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 15/100\n",
      "24/24 - 0s - loss: 0.0209 - mse: 0.0209 - mae: 0.0999 - mape: 1820847.6250 - 91ms/epoch - 4ms/step\n",
      "Epoch 16/100\n",
      "24/24 - 0s - loss: 0.0251 - mse: 0.0251 - mae: 0.1167 - mape: 1182568.7500 - 91ms/epoch - 4ms/step\n",
      "Epoch 17/100\n",
      "24/24 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1106 - mape: 1673725.1250 - 91ms/epoch - 4ms/step\n",
      "Epoch 18/100\n",
      "24/24 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1106 - mape: 1350166.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 19/100\n",
      "24/24 - 0s - loss: 0.0221 - mse: 0.0221 - mae: 0.1033 - mape: 1122480.6250 - 92ms/epoch - 4ms/step\n",
      "Epoch 20/100\n",
      "24/24 - 0s - loss: 0.0230 - mse: 0.0230 - mae: 0.1028 - mape: 773769.5625 - 92ms/epoch - 4ms/step\n",
      "Epoch 21/100\n",
      "24/24 - 0s - loss: 0.0234 - mse: 0.0234 - mae: 0.1105 - mape: 1155805.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 22/100\n",
      "24/24 - 0s - loss: 0.0236 - mse: 0.0236 - mae: 0.1078 - mape: 1389634.8750 - 102ms/epoch - 4ms/step\n",
      "Epoch 23/100\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1130 - mape: 1281943.0000 - 96ms/epoch - 4ms/step\n",
      "Epoch 24/100\n",
      "24/24 - 0s - loss: 0.0234 - mse: 0.0234 - mae: 0.1083 - mape: 1396140.0000 - 99ms/epoch - 4ms/step\n",
      "Epoch 25/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1027 - mape: 996155.8750 - 97ms/epoch - 4ms/step\n",
      "Epoch 26/100\n",
      "24/24 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1045 - mape: 2017243.3750 - 106ms/epoch - 4ms/step\n",
      "Epoch 27/100\n",
      "24/24 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1129 - mape: 1343497.3750 - 92ms/epoch - 4ms/step\n",
      "Epoch 28/100\n",
      "24/24 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1161 - mape: 1525025.2500 - 91ms/epoch - 4ms/step\n",
      "Epoch 29/100\n",
      "24/24 - 0s - loss: 0.0207 - mse: 0.0207 - mae: 0.1060 - mape: 778595.6250 - 101ms/epoch - 4ms/step\n",
      "Epoch 30/100\n",
      "24/24 - 0s - loss: 0.0264 - mse: 0.0264 - mae: 0.1140 - mape: 1973655.2500 - 105ms/epoch - 4ms/step\n",
      "Epoch 31/100\n",
      "24/24 - 0s - loss: 0.0222 - mse: 0.0222 - mae: 0.1036 - mape: 1211654.7500 - 98ms/epoch - 4ms/step\n",
      "Epoch 32/100\n",
      "24/24 - 0s - loss: 0.0186 - mse: 0.0186 - mae: 0.0941 - mape: 898522.5000 - 103ms/epoch - 4ms/step\n",
      "Epoch 33/100\n",
      "24/24 - 0s - loss: 0.0214 - mse: 0.0214 - mae: 0.1022 - mape: 1304625.1250 - 109ms/epoch - 5ms/step\n",
      "Epoch 34/100\n",
      "24/24 - 0s - loss: 0.0217 - mse: 0.0217 - mae: 0.1050 - mape: 962033.5625 - 101ms/epoch - 4ms/step\n",
      "Epoch 35/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.0999 - mape: 568022.9375 - 100ms/epoch - 4ms/step\n",
      "Epoch 36/100\n",
      "24/24 - 0s - loss: 0.0246 - mse: 0.0246 - mae: 0.1137 - mape: 2230864.7500 - 101ms/epoch - 4ms/step\n",
      "Epoch 37/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1068 - mape: 2360943.7500 - 101ms/epoch - 4ms/step\n",
      "Epoch 38/100\n",
      "24/24 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.0958 - mape: 805486.4375 - 99ms/epoch - 4ms/step\n",
      "Epoch 39/100\n",
      "24/24 - 0s - loss: 0.0194 - mse: 0.0194 - mae: 0.1005 - mape: 1109547.6250 - 102ms/epoch - 4ms/step\n",
      "Epoch 40/100\n",
      "24/24 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.0994 - mape: 1520511.6250 - 93ms/epoch - 4ms/step\n",
      "Epoch 41/100\n",
      "24/24 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.0965 - mape: 977581.4375 - 101ms/epoch - 4ms/step\n",
      "Epoch 42/100\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1005 - mape: 1414340.3750 - 102ms/epoch - 4ms/step\n",
      "Epoch 43/100\n",
      "24/24 - 0s - loss: 0.0201 - mse: 0.0201 - mae: 0.1007 - mape: 1881270.3750 - 94ms/epoch - 4ms/step\n",
      "Epoch 44/100\n",
      "24/24 - 0s - loss: 0.0197 - mse: 0.0197 - mae: 0.1063 - mape: 985323.5000 - 91ms/epoch - 4ms/step\n",
      "Epoch 45/100\n",
      "24/24 - 0s - loss: 0.0188 - mse: 0.0188 - mae: 0.0952 - mape: 1662860.2500 - 101ms/epoch - 4ms/step\n",
      "Epoch 46/100\n",
      "24/24 - 0s - loss: 0.0191 - mse: 0.0191 - mae: 0.1016 - mape: 709891.6875 - 102ms/epoch - 4ms/step\n",
      "Epoch 47/100\n",
      "24/24 - 0s - loss: 0.0181 - mse: 0.0181 - mae: 0.0965 - mape: 1048873.5000 - 93ms/epoch - 4ms/step\n",
      "Epoch 48/100\n",
      "24/24 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.0917 - mape: 778707.8125 - 101ms/epoch - 4ms/step\n",
      "Epoch 49/100\n",
      "24/24 - 0s - loss: 0.0197 - mse: 0.0197 - mae: 0.0985 - mape: 1492332.3750 - 102ms/epoch - 4ms/step\n",
      "Epoch 50/100\n",
      "24/24 - 0s - loss: 0.0172 - mse: 0.0172 - mae: 0.0992 - mape: 1259694.7500 - 94ms/epoch - 4ms/step\n",
      "Epoch 51/100\n",
      "24/24 - 0s - loss: 0.0171 - mse: 0.0171 - mae: 0.0961 - mape: 1333761.1250 - 100ms/epoch - 4ms/step\n",
      "Epoch 52/100\n",
      "24/24 - 0s - loss: 0.0173 - mse: 0.0173 - mae: 0.0963 - mape: 1216816.1250 - 101ms/epoch - 4ms/step\n",
      "Epoch 53/100\n",
      "24/24 - 0s - loss: 0.0174 - mse: 0.0174 - mae: 0.0973 - mape: 1144674.3750 - 102ms/epoch - 4ms/step\n",
      "Epoch 54/100\n",
      "24/24 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.0977 - mape: 1185613.8750 - 103ms/epoch - 4ms/step\n",
      "Epoch 55/100\n",
      "24/24 - 0s - loss: 0.0242 - mse: 0.0242 - mae: 0.1045 - mape: 1499145.1250 - 103ms/epoch - 4ms/step\n",
      "Epoch 56/100\n",
      "24/24 - 0s - loss: 0.0175 - mse: 0.0175 - mae: 0.1002 - mape: 1257670.5000 - 113ms/epoch - 5ms/step\n",
      "Epoch 57/100\n",
      "24/24 - 0s - loss: 0.0219 - mse: 0.0219 - mae: 0.1116 - mape: 1448477.2500 - 122ms/epoch - 5ms/step\n",
      "Epoch 58/100\n",
      "24/24 - 0s - loss: 0.0190 - mse: 0.0190 - mae: 0.1000 - mape: 662138.1250 - 132ms/epoch - 5ms/step\n",
      "Epoch 59/100\n",
      "24/24 - 0s - loss: 0.0162 - mse: 0.0162 - mae: 0.0946 - mape: 1163856.8750 - 123ms/epoch - 5ms/step\n",
      "Epoch 60/100\n",
      "24/24 - 0s - loss: 0.0180 - mse: 0.0180 - mae: 0.0968 - mape: 1146065.2500 - 130ms/epoch - 5ms/step\n",
      "Epoch 61/100\n",
      "24/24 - 0s - loss: 0.0176 - mse: 0.0176 - mae: 0.0983 - mape: 1922531.6250 - 132ms/epoch - 6ms/step\n",
      "Epoch 62/100\n",
      "24/24 - 0s - loss: 0.0165 - mse: 0.0165 - mae: 0.0960 - mape: 1267274.2500 - 125ms/epoch - 5ms/step\n",
      "Epoch 63/100\n",
      "24/24 - 0s - loss: 0.0161 - mse: 0.0161 - mae: 0.0887 - mape: 302036.7812 - 124ms/epoch - 5ms/step\n",
      "Epoch 64/100\n",
      "24/24 - 0s - loss: 0.0156 - mse: 0.0156 - mae: 0.0934 - mape: 1256050.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 65/100\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.0994 - mape: 1442232.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 66/100\n",
      "24/24 - 0s - loss: 0.0186 - mse: 0.0186 - mae: 0.0988 - mape: 1792790.8750 - 143ms/epoch - 6ms/step\n",
      "Epoch 67/100\n",
      "24/24 - 0s - loss: 0.0124 - mse: 0.0124 - mae: 0.0794 - mape: 946091.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 68/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.0989 - mape: 603344.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 69/100\n",
      "24/24 - 0s - loss: 0.0173 - mse: 0.0173 - mae: 0.0975 - mape: 923551.6250 - 119ms/epoch - 5ms/step\n",
      "Epoch 70/100\n",
      "24/24 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.0973 - mape: 1499585.3750 - 133ms/epoch - 6ms/step\n",
      "Epoch 71/100\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1126 - mape: 441699.7812 - 124ms/epoch - 5ms/step\n",
      "Epoch 72/100\n",
      "24/24 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.0886 - mape: 731914.1250 - 129ms/epoch - 5ms/step\n",
      "Epoch 73/100\n",
      "24/24 - 0s - loss: 0.0155 - mse: 0.0155 - mae: 0.0915 - mape: 784147.1250 - 132ms/epoch - 6ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "24/24 - 0s - loss: 0.0164 - mse: 0.0164 - mae: 0.0920 - mape: 268381.6250 - 124ms/epoch - 5ms/step\n",
      "Epoch 75/100\n",
      "24/24 - 0s - loss: 0.0185 - mse: 0.0185 - mae: 0.0954 - mape: 1129808.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 76/100\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0843 - mape: 882442.6875 - 117ms/epoch - 5ms/step\n",
      "Epoch 77/100\n",
      "24/24 - 0s - loss: 0.0206 - mse: 0.0206 - mae: 0.0927 - mape: 1105874.7500 - 119ms/epoch - 5ms/step\n",
      "Epoch 78/100\n",
      "24/24 - 0s - loss: 0.0180 - mse: 0.0180 - mae: 0.0959 - mape: 1463852.2500 - 120ms/epoch - 5ms/step\n",
      "Epoch 79/100\n",
      "24/24 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.0922 - mape: 727829.5625 - 129ms/epoch - 5ms/step\n",
      "Epoch 80/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1033 - mape: 1086225.7500 - 129ms/epoch - 5ms/step\n",
      "Epoch 81/100\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0863 - mape: 1524753.5000 - 125ms/epoch - 5ms/step\n",
      "Epoch 82/100\n",
      "24/24 - 0s - loss: 0.0199 - mse: 0.0199 - mae: 0.0984 - mape: 279958.7188 - 131ms/epoch - 5ms/step\n",
      "Epoch 83/100\n",
      "24/24 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.1017 - mape: 1513091.7500 - 130ms/epoch - 5ms/step\n",
      "Epoch 84/100\n",
      "24/24 - 0s - loss: 0.0196 - mse: 0.0196 - mae: 0.1014 - mape: 1816731.1250 - 126ms/epoch - 5ms/step\n",
      "Epoch 85/100\n",
      "24/24 - 0s - loss: 0.0184 - mse: 0.0184 - mae: 0.0885 - mape: 1376540.1250 - 130ms/epoch - 5ms/step\n",
      "Epoch 86/100\n",
      "24/24 - 0s - loss: 0.0171 - mse: 0.0171 - mae: 0.0976 - mape: 2094225.3750 - 122ms/epoch - 5ms/step\n",
      "Epoch 87/100\n",
      "24/24 - 0s - loss: 0.0137 - mse: 0.0137 - mae: 0.0843 - mape: 682203.2500 - 121ms/epoch - 5ms/step\n",
      "Epoch 88/100\n",
      "24/24 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.0972 - mape: 1263218.3750 - 124ms/epoch - 5ms/step\n",
      "Epoch 89/100\n",
      "24/24 - 0s - loss: 0.0168 - mse: 0.0168 - mae: 0.0908 - mape: 1680850.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 90/100\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0905 - mape: 1140126.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 91/100\n",
      "24/24 - 0s - loss: 0.0169 - mse: 0.0169 - mae: 0.0892 - mape: 451143.3125 - 121ms/epoch - 5ms/step\n",
      "Epoch 92/100\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0892 - mape: 874060.0625 - 122ms/epoch - 5ms/step\n",
      "Epoch 93/100\n",
      "24/24 - 0s - loss: 0.0144 - mse: 0.0144 - mae: 0.0905 - mape: 969346.0625 - 124ms/epoch - 5ms/step\n",
      "Epoch 94/100\n",
      "24/24 - 0s - loss: 0.0119 - mse: 0.0119 - mae: 0.0754 - mape: 396504.3125 - 122ms/epoch - 5ms/step\n",
      "Epoch 95/100\n",
      "24/24 - 0s - loss: 0.0134 - mse: 0.0134 - mae: 0.0847 - mape: 1552927.0000 - 118ms/epoch - 5ms/step\n",
      "Epoch 96/100\n",
      "24/24 - 0s - loss: 0.0193 - mse: 0.0193 - mae: 0.0916 - mape: 1278322.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 97/100\n",
      "24/24 - 0s - loss: 0.0169 - mse: 0.0169 - mae: 0.0858 - mape: 743073.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 98/100\n",
      "24/24 - 0s - loss: 0.0122 - mse: 0.0122 - mae: 0.0849 - mape: 1145978.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 99/100\n",
      "24/24 - 0s - loss: 0.0158 - mse: 0.0158 - mae: 0.0865 - mape: 989592.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 100/100\n",
      "24/24 - 0s - loss: 0.0113 - mse: 0.0113 - mae: 0.0778 - mape: 474013.1250 - 130ms/epoch - 5ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x29564d49f60>"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 6 include learning rate - looks better in plot, but slightly worse mape metric\n",
    "import keras\n",
    "from keras import layers\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, activation='relu', input_shape=(1, look_back), return_sequences=True))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(LSTM(50, activation='relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(1))\n",
    "\n",
    "opt = keras.optimizers.Adam(learning_rate=0.01)\n",
    "model.compile(optimizer=opt, loss='mse', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "id": "64411e02",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "25/25 - 3s - loss: 0.3781 - mse: 0.3781 - mae: 0.5734 - mape: 343366.5312 - 3s/epoch - 105ms/step\n",
      "Epoch 2/100\n",
      "25/25 - 0s - loss: 0.2531 - mse: 0.2531 - mae: 0.4595 - mape: 800469.9375 - 84ms/epoch - 3ms/step\n",
      "Epoch 3/100\n",
      "25/25 - 0s - loss: 0.0967 - mse: 0.0967 - mae: 0.2600 - mape: 2977747.5000 - 84ms/epoch - 3ms/step\n",
      "Epoch 4/100\n",
      "25/25 - 0s - loss: 0.0645 - mse: 0.0645 - mae: 0.2033 - mape: 2998732.5000 - 84ms/epoch - 3ms/step\n",
      "Epoch 5/100\n",
      "25/25 - 0s - loss: 0.0508 - mse: 0.0508 - mae: 0.1852 - mape: 4036331.2500 - 86ms/epoch - 3ms/step\n",
      "Epoch 6/100\n",
      "25/25 - 0s - loss: 0.0456 - mse: 0.0456 - mae: 0.1623 - mape: 6371222.0000 - 78ms/epoch - 3ms/step\n",
      "Epoch 7/100\n",
      "25/25 - 0s - loss: 0.0418 - mse: 0.0418 - mae: 0.1618 - mape: 5198816.5000 - 87ms/epoch - 3ms/step\n",
      "Epoch 8/100\n",
      "25/25 - 0s - loss: 0.0524 - mse: 0.0524 - mae: 0.1838 - mape: 3922587.5000 - 86ms/epoch - 3ms/step\n",
      "Epoch 9/100\n",
      "25/25 - 0s - loss: 0.0382 - mse: 0.0382 - mae: 0.1506 - mape: 5502281.5000 - 87ms/epoch - 3ms/step\n",
      "Epoch 10/100\n",
      "25/25 - 0s - loss: 0.0407 - mse: 0.0407 - mae: 0.1550 - mape: 5740967.5000 - 80ms/epoch - 3ms/step\n",
      "Epoch 11/100\n",
      "25/25 - 0s - loss: 0.0459 - mse: 0.0459 - mae: 0.1693 - mape: 4748164.0000 - 82ms/epoch - 3ms/step\n",
      "Epoch 12/100\n",
      "25/25 - 0s - loss: 0.0378 - mse: 0.0378 - mae: 0.1532 - mape: 4819262.0000 - 82ms/epoch - 3ms/step\n",
      "Epoch 13/100\n",
      "25/25 - 0s - loss: 0.0383 - mse: 0.0383 - mae: 0.1541 - mape: 5492315.5000 - 93ms/epoch - 4ms/step\n",
      "Epoch 14/100\n",
      "25/25 - 0s - loss: 0.0360 - mse: 0.0360 - mae: 0.1492 - mape: 4002857.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 15/100\n",
      "25/25 - 0s - loss: 0.0385 - mse: 0.0385 - mae: 0.1535 - mape: 5326598.0000 - 95ms/epoch - 4ms/step\n",
      "Epoch 16/100\n",
      "25/25 - 0s - loss: 0.0465 - mse: 0.0465 - mae: 0.1660 - mape: 4566505.5000 - 89ms/epoch - 4ms/step\n",
      "Epoch 17/100\n",
      "25/25 - 0s - loss: 0.0323 - mse: 0.0323 - mae: 0.1354 - mape: 4585293.0000 - 92ms/epoch - 4ms/step\n",
      "Epoch 18/100\n",
      "25/25 - 0s - loss: 0.0386 - mse: 0.0386 - mae: 0.1610 - mape: 5063165.0000 - 94ms/epoch - 4ms/step\n",
      "Epoch 19/100\n",
      "25/25 - 0s - loss: 0.0411 - mse: 0.0411 - mae: 0.1570 - mape: 5472541.5000 - 93ms/epoch - 4ms/step\n",
      "Epoch 20/100\n",
      "25/25 - 0s - loss: 0.0385 - mse: 0.0385 - mae: 0.1529 - mape: 5427185.5000 - 94ms/epoch - 4ms/step\n",
      "Epoch 21/100\n",
      "25/25 - 0s - loss: 0.0398 - mse: 0.0398 - mae: 0.1522 - mape: 5099969.0000 - 95ms/epoch - 4ms/step\n",
      "Epoch 22/100\n",
      "25/25 - 0s - loss: 0.0373 - mse: 0.0373 - mae: 0.1583 - mape: 4287452.0000 - 88ms/epoch - 4ms/step\n",
      "Epoch 23/100\n",
      "25/25 - 0s - loss: 0.0368 - mse: 0.0368 - mae: 0.1434 - mape: 4554268.0000 - 89ms/epoch - 4ms/step\n",
      "Epoch 24/100\n",
      "25/25 - 0s - loss: 0.0396 - mse: 0.0396 - mae: 0.1564 - mape: 3896858.5000 - 91ms/epoch - 4ms/step\n",
      "Epoch 25/100\n",
      "25/25 - 0s - loss: 0.0385 - mse: 0.0385 - mae: 0.1584 - mape: 4769866.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 26/100\n",
      "25/25 - 0s - loss: 0.0398 - mse: 0.0398 - mae: 0.1587 - mape: 4218882.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 27/100\n",
      "25/25 - 0s - loss: 0.0395 - mse: 0.0395 - mae: 0.1522 - mape: 5334770.0000 - 93ms/epoch - 4ms/step\n",
      "Epoch 28/100\n",
      "25/25 - 0s - loss: 0.0354 - mse: 0.0354 - mae: 0.1432 - mape: 4422370.5000 - 92ms/epoch - 4ms/step\n",
      "Epoch 29/100\n",
      "25/25 - 0s - loss: 0.0364 - mse: 0.0364 - mae: 0.1460 - mape: 5477005.5000 - 86ms/epoch - 3ms/step\n",
      "Epoch 30/100\n",
      "25/25 - 0s - loss: 0.0405 - mse: 0.0405 - mae: 0.1605 - mape: 6201566.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 31/100\n",
      "25/25 - 0s - loss: 0.0401 - mse: 0.0401 - mae: 0.1554 - mape: 5195775.0000 - 101ms/epoch - 4ms/step\n",
      "Epoch 32/100\n",
      "25/25 - 0s - loss: 0.0364 - mse: 0.0364 - mae: 0.1457 - mape: 4938398.0000 - 97ms/epoch - 4ms/step\n",
      "Epoch 33/100\n",
      "25/25 - 0s - loss: 0.0350 - mse: 0.0350 - mae: 0.1425 - mape: 5406443.5000 - 87ms/epoch - 3ms/step\n",
      "Epoch 34/100\n",
      "25/25 - 0s - loss: 0.0400 - mse: 0.0400 - mae: 0.1531 - mape: 5188844.0000 - 90ms/epoch - 4ms/step\n",
      "Epoch 35/100\n",
      "25/25 - 0s - loss: 0.0344 - mse: 0.0344 - mae: 0.1408 - mape: 6096640.5000 - 92ms/epoch - 4ms/step\n",
      "Epoch 36/100\n",
      "25/25 - 0s - loss: 0.0383 - mse: 0.0383 - mae: 0.1508 - mape: 4635380.0000 - 86ms/epoch - 3ms/step\n",
      "Epoch 37/100\n",
      "25/25 - 0s - loss: 0.0397 - mse: 0.0397 - mae: 0.1534 - mape: 5309085.5000 - 89ms/epoch - 4ms/step\n",
      "Epoch 38/100\n",
      "25/25 - 0s - loss: 0.0379 - mse: 0.0379 - mae: 0.1568 - mape: 4828983.0000 - 91ms/epoch - 4ms/step\n",
      "Epoch 39/100\n",
      "25/25 - 0s - loss: 0.0348 - mse: 0.0348 - mae: 0.1402 - mape: 5332405.5000 - 92ms/epoch - 4ms/step\n",
      "Epoch 40/100\n",
      "25/25 - 0s - loss: 0.0447 - mse: 0.0447 - mae: 0.1684 - mape: 5582345.0000 - 88ms/epoch - 4ms/step\n",
      "Epoch 41/100\n",
      "25/25 - 0s - loss: 0.0403 - mse: 0.0403 - mae: 0.1576 - mape: 4329874.0000 - 90ms/epoch - 4ms/step\n",
      "Epoch 42/100\n",
      "25/25 - 0s - loss: 0.0382 - mse: 0.0382 - mae: 0.1504 - mape: 5146136.5000 - 94ms/epoch - 4ms/step\n",
      "Epoch 43/100\n",
      "25/25 - 0s - loss: 0.0303 - mse: 0.0303 - mae: 0.1303 - mape: 5327540.5000 - 103ms/epoch - 4ms/step\n",
      "Epoch 44/100\n",
      "25/25 - 0s - loss: 0.0333 - mse: 0.0333 - mae: 0.1449 - mape: 4997728.5000 - 113ms/epoch - 5ms/step\n",
      "Epoch 45/100\n",
      "25/25 - 0s - loss: 0.0418 - mse: 0.0418 - mae: 0.1549 - mape: 5688215.5000 - 113ms/epoch - 5ms/step\n",
      "Epoch 46/100\n",
      "25/25 - 0s - loss: 0.0337 - mse: 0.0337 - mae: 0.1433 - mape: 4903672.0000 - 118ms/epoch - 5ms/step\n",
      "Epoch 47/100\n",
      "25/25 - 0s - loss: 0.0392 - mse: 0.0392 - mae: 0.1530 - mape: 4463425.0000 - 116ms/epoch - 5ms/step\n",
      "Epoch 48/100\n",
      "25/25 - 0s - loss: 0.0389 - mse: 0.0389 - mae: 0.1514 - mape: 5717759.5000 - 123ms/epoch - 5ms/step\n",
      "Epoch 49/100\n",
      "25/25 - 0s - loss: 0.0374 - mse: 0.0374 - mae: 0.1524 - mape: 5004598.5000 - 132ms/epoch - 5ms/step\n",
      "Epoch 50/100\n",
      "25/25 - 0s - loss: 0.0385 - mse: 0.0385 - mae: 0.1433 - mape: 5456528.0000 - 120ms/epoch - 5ms/step\n",
      "Epoch 51/100\n",
      "25/25 - 0s - loss: 0.0381 - mse: 0.0381 - mae: 0.1540 - mape: 5431528.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 52/100\n",
      "25/25 - 0s - loss: 0.0329 - mse: 0.0329 - mae: 0.1365 - mape: 5195669.0000 - 123ms/epoch - 5ms/step\n",
      "Epoch 53/100\n",
      "25/25 - 0s - loss: 0.0393 - mse: 0.0393 - mae: 0.1539 - mape: 4053397.5000 - 122ms/epoch - 5ms/step\n",
      "Epoch 54/100\n",
      "25/25 - 0s - loss: 0.0309 - mse: 0.0309 - mae: 0.1362 - mape: 4100597.5000 - 117ms/epoch - 5ms/step\n",
      "Epoch 55/100\n",
      "25/25 - 0s - loss: 0.0372 - mse: 0.0372 - mae: 0.1492 - mape: 4583512.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 56/100\n",
      "25/25 - 0s - loss: 0.0387 - mse: 0.0387 - mae: 0.1516 - mape: 5018475.5000 - 123ms/epoch - 5ms/step\n",
      "Epoch 57/100\n",
      "25/25 - 0s - loss: 0.0283 - mse: 0.0283 - mae: 0.1249 - mape: 4637599.5000 - 117ms/epoch - 5ms/step\n",
      "Epoch 58/100\n",
      "25/25 - 0s - loss: 0.0382 - mse: 0.0382 - mae: 0.1488 - mape: 5388470.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 59/100\n",
      "25/25 - 0s - loss: 0.0370 - mse: 0.0370 - mae: 0.1465 - mape: 4217010.0000 - 121ms/epoch - 5ms/step\n",
      "Epoch 60/100\n",
      "25/25 - 0s - loss: 0.0382 - mse: 0.0382 - mae: 0.1528 - mape: 4153849.5000 - 116ms/epoch - 5ms/step\n",
      "Epoch 61/100\n",
      "25/25 - 0s - loss: 0.0385 - mse: 0.0385 - mae: 0.1575 - mape: 5038579.5000 - 119ms/epoch - 5ms/step\n",
      "Epoch 62/100\n",
      "25/25 - 0s - loss: 0.0355 - mse: 0.0355 - mae: 0.1507 - mape: 4503168.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 63/100\n",
      "25/25 - 0s - loss: 0.0406 - mse: 0.0406 - mae: 0.1519 - mape: 6081389.5000 - 125ms/epoch - 5ms/step\n",
      "Epoch 64/100\n",
      "25/25 - 0s - loss: 0.0409 - mse: 0.0409 - mae: 0.1587 - mape: 4409442.0000 - 115ms/epoch - 5ms/step\n",
      "Epoch 65/100\n",
      "25/25 - 0s - loss: 0.0403 - mse: 0.0403 - mae: 0.1566 - mape: 4622898.0000 - 119ms/epoch - 5ms/step\n",
      "Epoch 66/100\n",
      "25/25 - 0s - loss: 0.0348 - mse: 0.0348 - mae: 0.1408 - mape: 6115556.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 67/100\n",
      "25/25 - 0s - loss: 0.0366 - mse: 0.0366 - mae: 0.1447 - mape: 5450330.0000 - 114ms/epoch - 5ms/step\n",
      "Epoch 68/100\n",
      "25/25 - 0s - loss: 0.0389 - mse: 0.0389 - mae: 0.1538 - mape: 5612964.0000 - 120ms/epoch - 5ms/step\n",
      "Epoch 69/100\n",
      "25/25 - 0s - loss: 0.0325 - mse: 0.0325 - mae: 0.1387 - mape: 4571775.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 70/100\n",
      "25/25 - 0s - loss: 0.0426 - mse: 0.0426 - mae: 0.1556 - mape: 4413124.5000 - 115ms/epoch - 5ms/step\n",
      "Epoch 71/100\n",
      "25/25 - 0s - loss: 0.0337 - mse: 0.0337 - mae: 0.1475 - mape: 4507827.0000 - 117ms/epoch - 5ms/step\n",
      "Epoch 72/100\n",
      "25/25 - 0s - loss: 0.0350 - mse: 0.0350 - mae: 0.1361 - mape: 4829899.0000 - 116ms/epoch - 5ms/step\n",
      "Epoch 73/100\n",
      "25/25 - 0s - loss: 0.0327 - mse: 0.0327 - mae: 0.1361 - mape: 5222878.5000 - 111ms/epoch - 4ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "25/25 - 0s - loss: 0.0376 - mse: 0.0376 - mae: 0.1478 - mape: 5177508.5000 - 117ms/epoch - 5ms/step\n",
      "Epoch 75/100\n",
      "25/25 - 0s - loss: 0.0418 - mse: 0.0418 - mae: 0.1570 - mape: 4829031.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 76/100\n",
      "25/25 - 0s - loss: 0.0386 - mse: 0.0386 - mae: 0.1514 - mape: 4711343.5000 - 114ms/epoch - 5ms/step\n",
      "Epoch 77/100\n",
      "25/25 - 0s - loss: 0.0351 - mse: 0.0351 - mae: 0.1515 - mape: 4282636.5000 - 118ms/epoch - 5ms/step\n",
      "Epoch 78/100\n",
      "25/25 - 0s - loss: 0.0403 - mse: 0.0403 - mae: 0.1517 - mape: 4764044.0000 - 121ms/epoch - 5ms/step\n",
      "Epoch 79/100\n",
      "25/25 - 0s - loss: 0.0381 - mse: 0.0381 - mae: 0.1475 - mape: 5460781.5000 - 116ms/epoch - 5ms/step\n",
      "Epoch 80/100\n",
      "25/25 - 0s - loss: 0.0355 - mse: 0.0355 - mae: 0.1422 - mape: 5602817.5000 - 119ms/epoch - 5ms/step\n",
      "Epoch 81/100\n",
      "25/25 - 0s - loss: 0.0310 - mse: 0.0310 - mae: 0.1313 - mape: 4993760.5000 - 116ms/epoch - 5ms/step\n",
      "Epoch 82/100\n",
      "25/25 - 0s - loss: 0.0358 - mse: 0.0358 - mae: 0.1420 - mape: 5025809.0000 - 113ms/epoch - 5ms/step\n",
      "Epoch 83/100\n",
      "25/25 - 0s - loss: 0.0348 - mse: 0.0348 - mae: 0.1457 - mape: 4606120.0000 - 118ms/epoch - 5ms/step\n",
      "Epoch 84/100\n",
      "25/25 - 0s - loss: 0.0390 - mse: 0.0390 - mae: 0.1505 - mape: 5012011.0000 - 115ms/epoch - 5ms/step\n",
      "Epoch 85/100\n",
      "25/25 - 0s - loss: 0.0355 - mse: 0.0355 - mae: 0.1398 - mape: 5209030.0000 - 118ms/epoch - 5ms/step\n",
      "Epoch 86/100\n",
      "25/25 - 0s - loss: 0.0364 - mse: 0.0364 - mae: 0.1476 - mape: 4268715.0000 - 126ms/epoch - 5ms/step\n",
      "Epoch 87/100\n",
      "25/25 - 0s - loss: 0.0370 - mse: 0.0370 - mae: 0.1423 - mape: 5410719.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 88/100\n",
      "25/25 - 0s - loss: 0.0350 - mse: 0.0350 - mae: 0.1442 - mape: 4564920.5000 - 113ms/epoch - 5ms/step\n",
      "Epoch 89/100\n",
      "25/25 - 0s - loss: 0.0367 - mse: 0.0367 - mae: 0.1469 - mape: 5289457.0000 - 117ms/epoch - 5ms/step\n",
      "Epoch 90/100\n",
      "25/25 - 0s - loss: 0.0354 - mse: 0.0354 - mae: 0.1470 - mape: 5178951.5000 - 111ms/epoch - 4ms/step\n",
      "Epoch 91/100\n",
      "25/25 - 0s - loss: 0.0355 - mse: 0.0355 - mae: 0.1482 - mape: 4437077.0000 - 117ms/epoch - 5ms/step\n",
      "Epoch 92/100\n",
      "25/25 - 0s - loss: 0.0290 - mse: 0.0290 - mae: 0.1294 - mape: 4801724.0000 - 118ms/epoch - 5ms/step\n",
      "Epoch 93/100\n",
      "25/25 - 0s - loss: 0.0367 - mse: 0.0367 - mae: 0.1509 - mape: 4883817.5000 - 116ms/epoch - 5ms/step\n",
      "Epoch 94/100\n",
      "25/25 - 0s - loss: 0.0334 - mse: 0.0334 - mae: 0.1415 - mape: 5181267.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 95/100\n",
      "25/25 - 0s - loss: 0.0342 - mse: 0.0342 - mae: 0.1406 - mape: 4368890.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 96/100\n",
      "25/25 - 0s - loss: 0.0357 - mse: 0.0357 - mae: 0.1451 - mape: 4157952.2500 - 116ms/epoch - 5ms/step\n",
      "Epoch 97/100\n",
      "25/25 - 0s - loss: 0.0378 - mse: 0.0378 - mae: 0.1487 - mape: 4627307.0000 - 122ms/epoch - 5ms/step\n",
      "Epoch 98/100\n",
      "25/25 - 0s - loss: 0.0378 - mse: 0.0378 - mae: 0.1495 - mape: 5449312.0000 - 114ms/epoch - 5ms/step\n",
      "Epoch 99/100\n",
      "25/25 - 0s - loss: 0.0388 - mse: 0.0388 - mae: 0.1504 - mape: 5084415.0000 - 117ms/epoch - 5ms/step\n",
      "Epoch 100/100\n",
      "25/25 - 0s - loss: 0.0416 - mse: 0.0416 - mae: 0.1545 - mape: 4531568.0000 - 113ms/epoch - 5ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2cc4ec8d3c0>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 7 increase dropouts for overfitting - worst of all of the above\n",
    "from tensorflow.keras.layers import Dropout\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, activation='relu', input_shape=(1, look_back), return_sequences=True, dropout = 0.5))\n",
    "#model.add(Dropout(0.2))\n",
    "model.add(LSTM(50, activation='relu', dropout = 0.5))\n",
    "#model.add(Dropout(0.2))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mse', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "b7cc1c26",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "24/24 - 3s - loss: 0.0896 - mse: 0.0896 - mae: 0.2363 - mape: 255994.2188 - 3s/epoch - 134ms/step\n",
      "Epoch 2/100\n",
      "24/24 - 0s - loss: 0.0551 - mse: 0.0551 - mae: 0.1710 - mape: 1087741.1250 - 110ms/epoch - 5ms/step\n",
      "Epoch 3/100\n",
      "24/24 - 0s - loss: 0.0330 - mse: 0.0330 - mae: 0.1222 - mape: 1963221.5000 - 100ms/epoch - 4ms/step\n",
      "Epoch 4/100\n",
      "24/24 - 0s - loss: 0.0308 - mse: 0.0308 - mae: 0.1310 - mape: 2175087.2500 - 92ms/epoch - 4ms/step\n",
      "Epoch 5/100\n",
      "24/24 - 0s - loss: 0.0295 - mse: 0.0295 - mae: 0.1205 - mape: 1697249.5000 - 100ms/epoch - 4ms/step\n",
      "Epoch 6/100\n",
      "24/24 - 0s - loss: 0.0302 - mse: 0.0302 - mae: 0.1292 - mape: 1340937.1250 - 92ms/epoch - 4ms/step\n",
      "Epoch 7/100\n",
      "24/24 - 0s - loss: 0.0280 - mse: 0.0280 - mae: 0.1219 - mape: 2376380.5000 - 92ms/epoch - 4ms/step\n",
      "Epoch 8/100\n",
      "24/24 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1218 - mape: 2289155.0000 - 102ms/epoch - 4ms/step\n",
      "Epoch 9/100\n",
      "24/24 - 0s - loss: 0.0280 - mse: 0.0280 - mae: 0.1175 - mape: 1647958.5000 - 102ms/epoch - 4ms/step\n",
      "Epoch 10/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1132 - mape: 1773459.2500 - 101ms/epoch - 4ms/step\n",
      "Epoch 11/100\n",
      "24/24 - 0s - loss: 0.0318 - mse: 0.0318 - mae: 0.1314 - mape: 1845243.3750 - 102ms/epoch - 4ms/step\n",
      "Epoch 12/100\n",
      "24/24 - 0s - loss: 0.0260 - mse: 0.0260 - mae: 0.1157 - mape: 2380967.0000 - 92ms/epoch - 4ms/step\n",
      "Epoch 13/100\n",
      "24/24 - 0s - loss: 0.0303 - mse: 0.0303 - mae: 0.1364 - mape: 2170510.5000 - 102ms/epoch - 4ms/step\n",
      "Epoch 14/100\n",
      "24/24 - 0s - loss: 0.0275 - mse: 0.0275 - mae: 0.1152 - mape: 2309630.0000 - 100ms/epoch - 4ms/step\n",
      "Epoch 15/100\n",
      "24/24 - 0s - loss: 0.0274 - mse: 0.0274 - mae: 0.1140 - mape: 1794049.2500 - 101ms/epoch - 4ms/step\n",
      "Epoch 16/100\n",
      "24/24 - 0s - loss: 0.0278 - mse: 0.0278 - mae: 0.1175 - mape: 2041001.1250 - 100ms/epoch - 4ms/step\n",
      "Epoch 17/100\n",
      "24/24 - 0s - loss: 0.0262 - mse: 0.0262 - mae: 0.1115 - mape: 2101470.0000 - 100ms/epoch - 4ms/step\n",
      "Epoch 18/100\n",
      "24/24 - 0s - loss: 0.0249 - mse: 0.0249 - mae: 0.1111 - mape: 2376862.5000 - 100ms/epoch - 4ms/step\n",
      "Epoch 19/100\n",
      "24/24 - 0s - loss: 0.0297 - mse: 0.0297 - mae: 0.1232 - mape: 1931058.2500 - 101ms/epoch - 4ms/step\n",
      "Epoch 20/100\n",
      "24/24 - 0s - loss: 0.0272 - mse: 0.0272 - mae: 0.1158 - mape: 1432681.2500 - 101ms/epoch - 4ms/step\n",
      "Epoch 21/100\n",
      "24/24 - 0s - loss: 0.0244 - mse: 0.0244 - mae: 0.1145 - mape: 1869814.8750 - 102ms/epoch - 4ms/step\n",
      "Epoch 22/100\n",
      "24/24 - 0s - loss: 0.0308 - mse: 0.0308 - mae: 0.1195 - mape: 1351348.5000 - 112ms/epoch - 5ms/step\n",
      "Epoch 23/100\n",
      "24/24 - 0s - loss: 0.0276 - mse: 0.0276 - mae: 0.1236 - mape: 1934242.1250 - 117ms/epoch - 5ms/step\n",
      "Epoch 24/100\n",
      "24/24 - 0s - loss: 0.0282 - mse: 0.0282 - mae: 0.1141 - mape: 1635757.3750 - 131ms/epoch - 5ms/step\n",
      "Epoch 25/100\n",
      "24/24 - 0s - loss: 0.0251 - mse: 0.0251 - mae: 0.1183 - mape: 1878056.2500 - 132ms/epoch - 6ms/step\n",
      "Epoch 26/100\n",
      "24/24 - 0s - loss: 0.0280 - mse: 0.0280 - mae: 0.1240 - mape: 2182360.5000 - 131ms/epoch - 5ms/step\n",
      "Epoch 27/100\n",
      "24/24 - 0s - loss: 0.0257 - mse: 0.0257 - mae: 0.1161 - mape: 1805644.5000 - 133ms/epoch - 6ms/step\n",
      "Epoch 28/100\n",
      "24/24 - 0s - loss: 0.0258 - mse: 0.0258 - mae: 0.1240 - mape: 2052507.5000 - 134ms/epoch - 6ms/step\n",
      "Epoch 29/100\n",
      "24/24 - 0s - loss: 0.0268 - mse: 0.0268 - mae: 0.1164 - mape: 1522327.7500 - 132ms/epoch - 5ms/step\n",
      "Epoch 30/100\n",
      "24/24 - 0s - loss: 0.0268 - mse: 0.0268 - mae: 0.1147 - mape: 1209661.1250 - 130ms/epoch - 5ms/step\n",
      "Epoch 31/100\n",
      "24/24 - 0s - loss: 0.0235 - mse: 0.0235 - mae: 0.1099 - mape: 1745027.5000 - 131ms/epoch - 5ms/step\n",
      "Epoch 32/100\n",
      "24/24 - 0s - loss: 0.0216 - mse: 0.0216 - mae: 0.1095 - mape: 1780818.8750 - 132ms/epoch - 5ms/step\n",
      "Epoch 33/100\n",
      "24/24 - 0s - loss: 0.0254 - mse: 0.0254 - mae: 0.1120 - mape: 1672967.6250 - 135ms/epoch - 6ms/step\n",
      "Epoch 34/100\n",
      "24/24 - 0s - loss: 0.0258 - mse: 0.0258 - mae: 0.1170 - mape: 1712051.6250 - 131ms/epoch - 5ms/step\n",
      "Epoch 35/100\n",
      "24/24 - 0s - loss: 0.0252 - mse: 0.0252 - mae: 0.1104 - mape: 1548724.1250 - 127ms/epoch - 5ms/step\n",
      "Epoch 36/100\n",
      "24/24 - 0s - loss: 0.0256 - mse: 0.0256 - mae: 0.1083 - mape: 1183502.8750 - 132ms/epoch - 5ms/step\n",
      "Epoch 37/100\n",
      "24/24 - 0s - loss: 0.0256 - mse: 0.0256 - mae: 0.1148 - mape: 1641216.7500 - 123ms/epoch - 5ms/step\n",
      "Epoch 38/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1037 - mape: 1655543.3750 - 139ms/epoch - 6ms/step\n",
      "Epoch 39/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1145 - mape: 1574730.3750 - 142ms/epoch - 6ms/step\n",
      "Epoch 40/100\n",
      "24/24 - 0s - loss: 0.0204 - mse: 0.0204 - mae: 0.1073 - mape: 1552942.7500 - 133ms/epoch - 6ms/step\n",
      "Epoch 41/100\n",
      "24/24 - 0s - loss: 0.0226 - mse: 0.0226 - mae: 0.1037 - mape: 1406354.3750 - 134ms/epoch - 6ms/step\n",
      "Epoch 42/100\n",
      "24/24 - 0s - loss: 0.0275 - mse: 0.0275 - mae: 0.1129 - mape: 1282260.1250 - 134ms/epoch - 6ms/step\n",
      "Epoch 43/100\n",
      "24/24 - 0s - loss: 0.0231 - mse: 0.0231 - mae: 0.1074 - mape: 1559524.6250 - 133ms/epoch - 6ms/step\n",
      "Epoch 44/100\n",
      "24/24 - 0s - loss: 0.0225 - mse: 0.0225 - mae: 0.1057 - mape: 1663264.7500 - 132ms/epoch - 5ms/step\n",
      "Epoch 45/100\n",
      "24/24 - 0s - loss: 0.0250 - mse: 0.0250 - mae: 0.1061 - mape: 1591184.3750 - 128ms/epoch - 5ms/step\n",
      "Epoch 46/100\n",
      "24/24 - 0s - loss: 0.0225 - mse: 0.0225 - mae: 0.1066 - mape: 1519117.6250 - 127ms/epoch - 5ms/step\n",
      "Epoch 47/100\n",
      "24/24 - 0s - loss: 0.0207 - mse: 0.0207 - mae: 0.1004 - mape: 1545204.8750 - 131ms/epoch - 5ms/step\n",
      "Epoch 48/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.1047 - mape: 1053434.1250 - 132ms/epoch - 6ms/step\n",
      "Epoch 49/100\n",
      "24/24 - 0s - loss: 0.0215 - mse: 0.0215 - mae: 0.1012 - mape: 1174265.3750 - 125ms/epoch - 5ms/step\n",
      "Epoch 50/100\n",
      "24/24 - 0s - loss: 0.0229 - mse: 0.0229 - mae: 0.1075 - mape: 1557710.5000 - 140ms/epoch - 6ms/step\n",
      "Epoch 51/100\n",
      "24/24 - 0s - loss: 0.0240 - mse: 0.0240 - mae: 0.1094 - mape: 1020507.4375 - 133ms/epoch - 6ms/step\n",
      "Epoch 52/100\n",
      "24/24 - 0s - loss: 0.0223 - mse: 0.0223 - mae: 0.1018 - mape: 1220981.8750 - 131ms/epoch - 5ms/step\n",
      "Epoch 53/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1056 - mape: 1474985.1250 - 132ms/epoch - 5ms/step\n",
      "Epoch 54/100\n",
      "24/24 - 0s - loss: 0.0221 - mse: 0.0221 - mae: 0.1054 - mape: 1240365.1250 - 130ms/epoch - 5ms/step\n",
      "Epoch 55/100\n",
      "24/24 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1034 - mape: 1061356.7500 - 132ms/epoch - 5ms/step\n",
      "Epoch 56/100\n",
      "24/24 - 0s - loss: 0.0193 - mse: 0.0193 - mae: 0.1049 - mape: 1188539.2500 - 132ms/epoch - 5ms/step\n",
      "Epoch 57/100\n",
      "24/24 - 0s - loss: 0.0221 - mse: 0.0221 - mae: 0.1055 - mape: 1382442.1250 - 124ms/epoch - 5ms/step\n",
      "Epoch 58/100\n",
      "24/24 - 0s - loss: 0.0186 - mse: 0.0186 - mae: 0.0939 - mape: 1070247.8750 - 131ms/epoch - 5ms/step\n",
      "Epoch 59/100\n",
      "24/24 - 0s - loss: 0.0193 - mse: 0.0193 - mae: 0.1057 - mape: 1497848.2500 - 132ms/epoch - 6ms/step\n",
      "Epoch 60/100\n",
      "24/24 - 0s - loss: 0.0179 - mse: 0.0179 - mae: 0.0977 - mape: 1237984.1250 - 132ms/epoch - 5ms/step\n",
      "Epoch 61/100\n",
      "24/24 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1043 - mape: 805158.6875 - 124ms/epoch - 5ms/step\n",
      "Epoch 62/100\n",
      "24/24 - 0s - loss: 0.0212 - mse: 0.0212 - mae: 0.1012 - mape: 1127148.3750 - 132ms/epoch - 6ms/step\n",
      "Epoch 63/100\n",
      "24/24 - 0s - loss: 0.0207 - mse: 0.0207 - mae: 0.1010 - mape: 981373.6875 - 125ms/epoch - 5ms/step\n",
      "Epoch 64/100\n",
      "24/24 - 0s - loss: 0.0206 - mse: 0.0206 - mae: 0.1037 - mape: 1365415.7500 - 120ms/epoch - 5ms/step\n",
      "Epoch 65/100\n",
      "24/24 - 0s - loss: 0.0210 - mse: 0.0210 - mae: 0.1033 - mape: 906839.3125 - 125ms/epoch - 5ms/step\n",
      "Epoch 66/100\n",
      "24/24 - 0s - loss: 0.0208 - mse: 0.0208 - mae: 0.0991 - mape: 1169649.7500 - 131ms/epoch - 5ms/step\n",
      "Epoch 67/100\n",
      "24/24 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.1037 - mape: 2480659.7500 - 123ms/epoch - 5ms/step\n",
      "Epoch 68/100\n",
      "24/24 - 0s - loss: 0.0218 - mse: 0.0218 - mae: 0.1026 - mape: 1495655.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 69/100\n",
      "24/24 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.1018 - mape: 899488.0625 - 145ms/epoch - 6ms/step\n",
      "Epoch 70/100\n",
      "24/24 - 0s - loss: 0.0160 - mse: 0.0160 - mae: 0.0933 - mape: 1550270.2500 - 132ms/epoch - 5ms/step\n",
      "Epoch 71/100\n",
      "24/24 - 0s - loss: 0.0192 - mse: 0.0192 - mae: 0.1002 - mape: 711928.0000 - 131ms/epoch - 5ms/step\n",
      "Epoch 72/100\n",
      "24/24 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1016 - mape: 1176368.1250 - 124ms/epoch - 5ms/step\n",
      "Epoch 73/100\n",
      "24/24 - 0s - loss: 0.0187 - mse: 0.0187 - mae: 0.0962 - mape: 1466361.5000 - 122ms/epoch - 5ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 74/100\n",
      "24/24 - 0s - loss: 0.0202 - mse: 0.0202 - mae: 0.0961 - mape: 1063276.7500 - 121ms/epoch - 5ms/step\n",
      "Epoch 75/100\n",
      "24/24 - 0s - loss: 0.0154 - mse: 0.0154 - mae: 0.0897 - mape: 998574.8750 - 117ms/epoch - 5ms/step\n",
      "Epoch 76/100\n",
      "24/24 - 0s - loss: 0.0194 - mse: 0.0194 - mae: 0.1020 - mape: 1694354.7500 - 122ms/epoch - 5ms/step\n",
      "Epoch 77/100\n",
      "24/24 - 0s - loss: 0.0151 - mse: 0.0151 - mae: 0.0897 - mape: 500869.0000 - 128ms/epoch - 5ms/step\n",
      "Epoch 78/100\n",
      "24/24 - 0s - loss: 0.0220 - mse: 0.0220 - mae: 0.1033 - mape: 1298904.2500 - 132ms/epoch - 6ms/step\n",
      "Epoch 79/100\n",
      "24/24 - 0s - loss: 0.0186 - mse: 0.0186 - mae: 0.0946 - mape: 1199875.0000 - 131ms/epoch - 5ms/step\n",
      "Epoch 80/100\n",
      "24/24 - 0s - loss: 0.0203 - mse: 0.0203 - mae: 0.1011 - mape: 1126911.3750 - 124ms/epoch - 5ms/step\n",
      "Epoch 81/100\n",
      "24/24 - 0s - loss: 0.0198 - mse: 0.0198 - mae: 0.1026 - mape: 719599.6250 - 122ms/epoch - 5ms/step\n",
      "Epoch 82/100\n",
      "24/24 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.0904 - mape: 1410223.6250 - 119ms/epoch - 5ms/step\n",
      "Epoch 83/100\n",
      "24/24 - 0s - loss: 0.0193 - mse: 0.0193 - mae: 0.1019 - mape: 1184437.3750 - 121ms/epoch - 5ms/step\n",
      "Epoch 84/100\n",
      "24/24 - 0s - loss: 0.0189 - mse: 0.0189 - mae: 0.0999 - mape: 1403473.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 85/100\n",
      "24/24 - 0s - loss: 0.0187 - mse: 0.0187 - mae: 0.0916 - mape: 642726.9375 - 128ms/epoch - 5ms/step\n",
      "Epoch 86/100\n",
      "24/24 - 0s - loss: 0.0178 - mse: 0.0178 - mae: 0.0926 - mape: 1648195.6250 - 130ms/epoch - 5ms/step\n",
      "Epoch 87/100\n",
      "24/24 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.0998 - mape: 927702.0000 - 131ms/epoch - 5ms/step\n",
      "Epoch 88/100\n",
      "24/24 - 0s - loss: 0.0195 - mse: 0.0195 - mae: 0.0989 - mape: 863306.5000 - 121ms/epoch - 5ms/step\n",
      "Epoch 89/100\n",
      "24/24 - 0s - loss: 0.0155 - mse: 0.0155 - mae: 0.0905 - mape: 1557999.6250 - 133ms/epoch - 6ms/step\n",
      "Epoch 90/100\n",
      "24/24 - 0s - loss: 0.0173 - mse: 0.0173 - mae: 0.0949 - mape: 1188281.2500 - 123ms/epoch - 5ms/step\n",
      "Epoch 91/100\n",
      "24/24 - 0s - loss: 0.0152 - mse: 0.0152 - mae: 0.0908 - mape: 1193522.8750 - 124ms/epoch - 5ms/step\n",
      "Epoch 92/100\n",
      "24/24 - 0s - loss: 0.0167 - mse: 0.0167 - mae: 0.0910 - mape: 1157240.5000 - 120ms/epoch - 5ms/step\n",
      "Epoch 93/100\n",
      "24/24 - 0s - loss: 0.0163 - mse: 0.0163 - mae: 0.0886 - mape: 253829.0469 - 133ms/epoch - 6ms/step\n",
      "Epoch 94/100\n",
      "24/24 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0915 - mape: 1131798.7500 - 124ms/epoch - 5ms/step\n",
      "Epoch 95/100\n",
      "24/24 - 0s - loss: 0.0150 - mse: 0.0150 - mae: 0.0902 - mape: 1057504.1250 - 122ms/epoch - 5ms/step\n",
      "Epoch 96/100\n",
      "24/24 - 0s - loss: 0.0131 - mse: 0.0131 - mae: 0.0871 - mape: 1136905.3750 - 128ms/epoch - 5ms/step\n",
      "Epoch 97/100\n",
      "24/24 - 0s - loss: 0.0177 - mse: 0.0177 - mae: 0.0963 - mape: 1365465.6250 - 132ms/epoch - 5ms/step\n",
      "Epoch 98/100\n",
      "24/24 - 0s - loss: 0.0148 - mse: 0.0148 - mae: 0.0928 - mape: 1555136.8750 - 122ms/epoch - 5ms/step\n",
      "Epoch 99/100\n",
      "24/24 - 0s - loss: 0.0147 - mse: 0.0147 - mae: 0.0873 - mape: 937025.7500 - 132ms/epoch - 6ms/step\n",
      "Epoch 100/100\n",
      "24/24 - 0s - loss: 0.0172 - mse: 0.0172 - mae: 0.0899 - mape: 1118566.2500 - 131ms/epoch - 5ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x295695ea200>"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# option 8 different dropouts for overfitting - better\n",
    "from tensorflow.keras.layers import Dropout\n",
    "\n",
    "# create and fit the LSTM network\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, activation='relu', input_shape=(1, look_back), return_sequences=True))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(LSTM(50, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mse', metrics=['mse', 'mae', 'mape'])\n",
    "model.fit(trainX, trainY, epochs=100, batch_size=4, verbose=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db39fcc2",
   "metadata": {},
   "source": [
    "### predict future values based on trained model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "825ca5dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3/3 [==============================] - 0s 2ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "Train Score: 12036.71 RMSE\n",
      "Test Score: 15553.57 RMSE\n"
     ]
    }
   ],
   "source": [
    "# predict on given data\n",
    "trainPredict = model.predict(trainX)\n",
    "testPredict = model.predict(testX)\n",
    "# invert predictions\n",
    "trainPredict = scaler.inverse_transform(trainPredict)\n",
    "trainY = scaler.inverse_transform([trainY])\n",
    "testPredict = scaler.inverse_transform(testPredict)\n",
    "testY = scaler.inverse_transform([testY])\n",
    "# calculate root mean squared error\n",
    "trainScore = np.sqrt(mean_squared_error(trainY[0], trainPredict[:,0]))\n",
    "print('Train Score: %.2f RMSE' % (trainScore))\n",
    "testScore = np.sqrt(mean_squared_error(testY[0], testPredict[:,0]))\n",
    "print('Test Score: %.2f RMSE' % (testScore))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "b8ec55f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 16ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 18ms/step\n",
      "1/1 [==============================] - 0s 11ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 25ms/step\n",
      "1/1 [==============================] - 0s 12ms/step\n",
      "1/1 [==============================] - 0s 19ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 32ms/step\n",
      "1/1 [==============================] - 0s 9ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "1/1 [==============================] - 0s 31ms/step\n",
      "1/1 [==============================] - 0s 21ms/step\n",
      "1/1 [==============================] - 0s 13ms/step\n",
      "1/1 [==============================] - 0s 23ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 24ms/step\n",
      "1/1 [==============================] - 0s 22ms/step\n",
      "Future Predictions:\n",
      "[[53501.266]\n",
      " [49383.76 ]\n",
      " [47425.02 ]\n",
      " [54958.594]\n",
      " [59159.297]\n",
      " [50186.668]\n",
      " [63083.773]\n",
      " [59131.227]\n",
      " [62012.543]\n",
      " [73973.3  ]\n",
      " [55129.277]\n",
      " [62994.277]\n",
      " [74648.35 ]\n",
      " [63030.83 ]\n",
      " [67757.05 ]\n",
      " [76454.47 ]\n",
      " [57802.844]\n",
      " [77179.98 ]\n",
      " [78815.86 ]\n",
      " [47989.95 ]\n",
      " [64833.82 ]\n",
      " [71694.836]\n",
      " [52767.227]\n",
      " [71235.79 ]\n",
      " [64630.383]\n",
      " [59928.09 ]]\n"
     ]
    }
   ],
   "source": [
    "# predict future values\n",
    "\n",
    "# Prepare the last sequence from the dataset\n",
    "last_sequence = dataset[-look_back:]\n",
    "\n",
    "# Initialize a list to store the predictions\n",
    "future_predictions = []\n",
    "\n",
    "# Iterate to predict\n",
    "for _ in range(26):\n",
    "    # Reshape the last sequence to the shape [samples, time steps, features]\n",
    "    input_seq = np.reshape(last_sequence, (1, 1, look_back))\n",
    "    \n",
    "    # Predict the next value\n",
    "    next_value = model.predict(input_seq)\n",
    "    \n",
    "    # Append the predicted value to the predictions list\n",
    "    future_predictions.append(next_value[0, 0])\n",
    "    \n",
    "    # Update the last sequence by removing the first value and appending the predicted value\n",
    "    last_sequence = np.append(last_sequence[1:], next_value)\n",
    "    \n",
    "    # Reshape back to (look_back,) for the next iteration\n",
    "    last_sequence = last_sequence.reshape((look_back,))\n",
    "\n",
    "# Convert predictions to a numpy array and scale back to the original scale\n",
    "future_predictions = np.array(future_predictions).reshape(-1, 1)\n",
    "future_predictions_original_scale = scaler.inverse_transform(future_predictions)\n",
    "\n",
    "# Print the future predictions in the original scale\n",
    "print(\"Future Predictions:\")\n",
    "print(future_predictions_original_scale)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "id": "60d33489",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from datetime import timedelta\n",
    "\n",
    "dates = lstmdatainput[\"OrderDate\"]\n",
    "last_date = dates.iloc[-1]\n",
    "future_dates = [last_date + timedelta(weeks=i) for i in range(1, 27)]\n",
    "future_dates = pd.to_datetime(future_dates, format='%Y-%m-%d')\n",
    "# Combine original dates with future dates\n",
    "extended_dates = pd.Series(np.concatenate([dates.values, future_dates]))\n",
    "extended_dates = pd.to_datetime(extended_dates, format='%Y-%m-%d')\n",
    "\n",
    "# Create a new array to hold the dataset including the future predictions\n",
    "extended_dataset = np.append(dataset, future_predictions, axis=0)\n",
    "\n",
    "# Inverse transform the extended dataset\n",
    "extended_dataset_original_scale = scaler.inverse_transform(extended_dataset)\n",
    "\n",
    "# Create an array to hold the future predictions for plotting\n",
    "futurePredictPlot = np.empty_like(extended_dataset)\n",
    "futurePredictPlot[:, :] = np.nan\n",
    "futurePredictPlot[len(dataset):, :] = future_predictions_original_scale\n",
    "\n",
    "# Shift train predictions for plotting\n",
    "trainPredictPlot = np.empty_like(extended_dataset)\n",
    "trainPredictPlot[:, :] = np.nan\n",
    "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict\n",
    "\n",
    "# Shift test predictions for plotting\n",
    "testPredictPlot = np.empty_like(extended_dataset)\n",
    "testPredictPlot[:, :] = np.nan\n",
    "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(dataset)-1, :] = testPredict\n",
    "\n",
    "# Combine extended dataset with dates\n",
    "extended_dataset_with_dates = pd.DataFrame({\n",
    "    'Date': extended_dates,\n",
    "    'Revenue': extended_dataset_original_scale.flatten(),\n",
    "    'Train' : trainPredictPlot.flatten(),\n",
    "    'Test' : testPredictPlot.flatten(),\n",
    "    'Predict' : futurePredictPlot.flatten(),\n",
    "})\n",
    "\n",
    "# Plot baseline and predictions\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.plot(extended_dataset_with_dates[\"Date\"], extended_dataset_with_dates[\"Revenue\"], label='Original Data')\n",
    "plt.plot(extended_dataset_with_dates[\"Date\"], extended_dataset_with_dates[\"Train\"], label='Train Predictions')\n",
    "plt.plot(extended_dataset_with_dates[\"Date\"], extended_dataset_with_dates[\"Test\"], label='Test Predictions')\n",
    "plt.plot(extended_dataset_with_dates[\"Date\"], extended_dataset_with_dates[\"Predict\"], label='Future Predictions')\n",
    "plt.legend()\n",
    "plt.title(\"2024 - Weekly Predictions of Revenue for Customers in Cluster 3\")\n",
    "plt.xlabel('Date (weeks)')\n",
    "plt.ylabel('Revenue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "2a93f097",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 156 entries, 0 to 155\n",
      "Data columns (total 5 columns):\n",
      " #   Column   Non-Null Count  Dtype         \n",
      "---  ------   --------------  -----         \n",
      " 0   Date     156 non-null    datetime64[ns]\n",
      " 1   Revenue  156 non-null    float32       \n",
      " 2   Train    93 non-null     float32       \n",
      " 3   Test     15 non-null     float32       \n",
      " 4   Predict  26 non-null     float32       \n",
      "dtypes: datetime64[ns](1), float32(4)\n",
      "memory usage: 3.8 KB\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Predict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>2024-06-24</td>\n",
       "      <td>53501.265625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>2024-07-01</td>\n",
       "      <td>49383.761719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>2024-07-08</td>\n",
       "      <td>47425.019531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>54958.593750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134</th>\n",
       "      <td>2024-07-22</td>\n",
       "      <td>59159.296875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>2024-07-29</td>\n",
       "      <td>50186.667969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>63083.773438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>137</th>\n",
       "      <td>2024-08-12</td>\n",
       "      <td>59131.226562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>138</th>\n",
       "      <td>2024-08-19</td>\n",
       "      <td>62012.542969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>139</th>\n",
       "      <td>2024-08-26</td>\n",
       "      <td>73973.296875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>2024-09-02</td>\n",
       "      <td>55129.277344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>2024-09-09</td>\n",
       "      <td>62994.277344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>142</th>\n",
       "      <td>2024-09-16</td>\n",
       "      <td>74648.351562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>2024-09-23</td>\n",
       "      <td>63030.828125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>2024-09-30</td>\n",
       "      <td>67757.046875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>2024-10-07</td>\n",
       "      <td>76454.468750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>2024-10-14</td>\n",
       "      <td>57802.843750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>2024-10-21</td>\n",
       "      <td>77179.976562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>2024-10-28</td>\n",
       "      <td>78815.859375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>2024-11-04</td>\n",
       "      <td>47989.949219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>2024-11-11</td>\n",
       "      <td>64833.820312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>151</th>\n",
       "      <td>2024-11-18</td>\n",
       "      <td>71694.835938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>2024-11-25</td>\n",
       "      <td>52767.226562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2024-12-02</td>\n",
       "      <td>71235.789062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>154</th>\n",
       "      <td>2024-12-09</td>\n",
       "      <td>64630.382812</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>155</th>\n",
       "      <td>2024-12-16</td>\n",
       "      <td>59928.089844</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Date       Predict\n",
       "130 2024-06-24  53501.265625\n",
       "131 2024-07-01  49383.761719\n",
       "132 2024-07-08  47425.019531\n",
       "133 2024-07-15  54958.593750\n",
       "134 2024-07-22  59159.296875\n",
       "135 2024-07-29  50186.667969\n",
       "136 2024-08-05  63083.773438\n",
       "137 2024-08-12  59131.226562\n",
       "138 2024-08-19  62012.542969\n",
       "139 2024-08-26  73973.296875\n",
       "140 2024-09-02  55129.277344\n",
       "141 2024-09-09  62994.277344\n",
       "142 2024-09-16  74648.351562\n",
       "143 2024-09-23  63030.828125\n",
       "144 2024-09-30  67757.046875\n",
       "145 2024-10-07  76454.468750\n",
       "146 2024-10-14  57802.843750\n",
       "147 2024-10-21  77179.976562\n",
       "148 2024-10-28  78815.859375\n",
       "149 2024-11-04  47989.949219\n",
       "150 2024-11-11  64833.820312\n",
       "151 2024-11-18  71694.835938\n",
       "152 2024-11-25  52767.226562\n",
       "153 2024-12-02  71235.789062\n",
       "154 2024-12-09  64630.382812\n",
       "155 2024-12-16  59928.089844"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "extended_dataset_with_dates.info()\n",
    "filtered_extended_dataset_with_dates = extended_dataset_with_dates.dropna(subset=['Predict'])\n",
    "filtered_extended_dataset_with_dates = filtered_extended_dataset_with_dates.drop(['Revenue','Train', 'Test'], axis=1)\n",
    "filtered_extended_dataset_with_dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "d3d02ca2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#extended_dataset_with_dates.to_csv(\"2024predictionsCluster3sample.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "4e6d79fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# shift train predictions for plotting\n",
    "trainPredictPlot = np.empty_like(dataset)\n",
    "trainPredictPlot[:, :] = np.nan\n",
    "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict\n",
    "# shift test predictions for plotting\n",
    "testPredictPlot = np.empty_like(dataset)\n",
    "testPredictPlot[:, :] = np.nan\n",
    "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(dataset)-1, :] = testPredict\n",
    "# plot baseline and predictions\n",
    "plt.plot(scaler.inverse_transform(dataset))\n",
    "plt.plot(trainPredictPlot)\n",
    "plt.plot(testPredictPlot)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "286b751c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     CustomerID  cluster\n",
      "362    10255544        2\n",
      "571    10403731        2\n",
      "629    10446652        2\n",
      "660    10465952        2\n",
      "1748     150085        2\n",
      "\n",
      "   OrderDate    NetRevenue\n",
      "0 2021-12-27   3061.350785\n",
      "1 2022-01-03  20784.907068\n",
      "2 2022-01-10  12727.215969\n",
      "3 2022-01-17  14527.650524\n",
      "4 2022-01-24  11662.307592\n",
      "\n",
      "[[40825.246]\n",
      " [50121.38 ]\n",
      " [41858.62 ]\n",
      " [52709.746]\n",
      " [46157.89 ]\n",
      " [45872.492]\n",
      " [49109.125]\n",
      " [46092.684]\n",
      " [48680.23 ]\n",
      " [50243.508]\n",
      " [49964.188]\n",
      " [51289.08 ]\n",
      " [46590.5  ]\n",
      " [53914.34 ]\n",
      " [44794.445]\n",
      " [48293.91 ]\n",
      " [51784.863]\n",
      " [43844.97 ]\n",
      " [47306.305]\n",
      " [52181.86 ]\n",
      " [52234.82 ]\n",
      " [45398.805]\n",
      " [49176.777]\n",
      " [49328.08 ]]\n"
     ]
    }
   ],
   "source": [
    "# proposed relevant datasets for dashboard\n",
    "# customer-cluster assignments\n",
    "# predictions for upcoming 6 months as dataframe"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}