abegehr/Coursera_ADSC_covid19-week4-simple-lstm.ipynb

## Coursera_ADSC_covid19-week4-simple-lstm.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple LSTM\n",
    "Using a simple LSTM recurrent neural network to predict cases and fatalities"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports and Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/kaggle/input/covid19-global-forecasting-week-4/train.csv\n",
      "/kaggle/input/covid19-global-forecasting-week-4/submission.csv\n",
      "/kaggle/input/covid19-global-forecasting-week-4/test.csv\n"
     ]
    }
   ],
   "source": [
    "# This Python 3 environment comes with many helpful analytics libraries installed\n",
    "# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python\n",
    "# For example, here's several helpful packages to load in \n",
    "\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "\n",
    "# Input data files are available in the \"../input/\" directory.\n",
    "# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n",
    "\n",
    "import os\n",
    "for dirname, _, filenames in os.walk('/kaggle/input'):\n",
    "    for filename in filenames:\n",
    "        print(os.path.join(dirname, filename))\n",
    "\n",
    "# Any results you write to the current directory are saved as output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM\n",
    "import keras.backend as K\n",
    "import tensorflow as tf\n",
    "from sklearn.metrics import mean_squared_log_error\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "test = pd.read_csv(\"../input/covid19-global-forecasting-week-4/test.csv\", parse_dates=[\"Date\"])\n",
    "train = pd.read_csv(\"../input/covid19-global-forecasting-week-4/train.csv\")\n",
    "submission = pd.read_csv(\"../input/covid19-global-forecasting-week-4/submission.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LSTM Model and Helpers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Id</th>\n",
       "      <th>Province_State</th>\n",
       "      <th>Country_Region</th>\n",
       "      <th>Date</th>\n",
       "      <th>ConfirmedCases</th>\n",
       "      <th>Fatalities</th>\n",
       "      <th>geo_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>2020-01-22</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Afghanistan_nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>2020-01-23</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Afghanistan_nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>2020-01-24</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Afghanistan_nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>2020-01-25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Afghanistan_nan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Afghanistan</td>\n",
       "      <td>2020-01-26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Afghanistan_nan</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id Province_State Country_Region        Date  ConfirmedCases  Fatalities  \\\n",
       "0   1            NaN    Afghanistan  2020-01-22             0.0         0.0   \n",
       "1   2            NaN    Afghanistan  2020-01-23             0.0         0.0   \n",
       "2   3            NaN    Afghanistan  2020-01-24             0.0         0.0   \n",
       "3   4            NaN    Afghanistan  2020-01-25             0.0         0.0   \n",
       "4   5            NaN    Afghanistan  2020-01-26             0.0         0.0   \n",
       "\n",
       "            geo_id  \n",
       "0  Afghanistan_nan  \n",
       "1  Afghanistan_nan  \n",
       "2  Afghanistan_nan  \n",
       "3  Afghanistan_nan  \n",
       "4  Afghanistan_nan  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# generate key for each entitiy to predict for\n",
    "train['geo_id'] = train['Country_Region'].astype(str) + '_' + train['Province_State'].astype(str)\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_entity_data_split(geo_id, train_split_factor=1.0):\n",
    "    data = train[train[\"geo_id\"] == geo_id]\n",
    "    country = data[\"Country_Region\"].unique()[0]\n",
    "    province = data[\"Province_State\"].unique()[0]\n",
    "    \n",
    "    case = data[\"ConfirmedCases\"].to_numpy()\n",
    "    fat = data[\"Fatalities\"].to_numpy()\n",
    "    case = case.reshape((len(case), 1))\n",
    "    fat = fat.reshape((len(fat), 1))\n",
    "    \n",
    "    train_size = int(len(data) * train_split_factor)\n",
    "    test_size = len(data) - train_size\n",
    "    \n",
    "    case_train, case_test = case[0:train_size,:], case[train_size:len(data),:]\n",
    "    fat_train, fat_test = fat[0:train_size,:], fat[train_size:len(data),:]\n",
    "    \n",
    "    return train_size, test_size, case_train, case_test, fat_train, fat_test\n",
    "\n",
    "def create_dataset_for_lstm(data, look_back=1.0):\n",
    "    # reshape into X=t and Y=t+1\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(data)-look_back-1):\n",
    "        a = data[i:(i+look_back), 0]\n",
    "        dataX.append(a)\n",
    "        dataY.append(data[i + look_back, 0])\n",
    "    X = np.array(dataX)\n",
    "    Y = np.array(dataY)\n",
    "    \n",
    "    # reshape input to be [samples, time steps, features]\n",
    "    X = np.reshape(X, (X.shape[0], 1, X.shape[1]))\n",
    "    \n",
    "    return X, Y\n",
    "\n",
    "def calculate_loss(Y, pred):\n",
    "    return np.sqrt(mean_squared_log_error(Y[0], pred[:,0]))\n",
    "\n",
    "'''\n",
    "def rmsle_K(y, y0):\n",
    "    return K.sqrt(K.mean(K.square(tf.math.log1p(y) - tf.math.log1p(y0))))\n",
    "'''\n",
    "\n",
    "def train_lstm(train, test=[], look_back=1, epochs=10, batch_size=1, verbose=0):\n",
    "    # scale\n",
    "    scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "    train_scaled = scaler.fit_transform(train)\n",
    "    test_scaled = scaler.transform(test) if (len(test) > 0) else []\n",
    "    \n",
    "    # prep data\n",
    "    trainX, trainY = create_dataset_for_lstm(train_scaled, look_back=look_back)\n",
    "    testX, testY = (create_dataset_for_lstm(test_scaled, look_back=look_back)) if (len(test) > 0) else ([], [])\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",
    "    model.compile(loss='mean_squared_error', optimizer='adam')\n",
    "    model.fit(trainX, trainY, epochs=epochs, batch_size=batch_size, verbose=verbose)\n",
    "    \n",
    "    # make predictions\n",
    "    trainPredict = model.predict(trainX).clip(0)\n",
    "    testPredict = model.predict(testX).clip(0) if (len(test) > 0) else []\n",
    "    \n",
    "    # invert predictions\n",
    "    trainPredict = scaler.inverse_transform(trainPredict)\n",
    "    trainY = scaler.inverse_transform([trainY])\n",
    "    testPredict = scaler.inverse_transform(testPredict) if (len(testPredict) > 0) else []\n",
    "    testY = scaler.inverse_transform([testY]) if (len(testY) > 0) else []\n",
    "    \n",
    "    # calculate RMSLE\n",
    "    trainScore = calculate_loss(trainY, trainPredict)\n",
    "    testScore = calculate_loss(testY, testPredict) if (len(test) > 0) else 0\n",
    "    if (verbose):\n",
    "        print('Train Score: %.2f RMSLE' % (trainScore))\n",
    "        print('Test Score: %.2f RMSLE' % (testScore))\n",
    "    \n",
    "    return {\n",
    "        \"model\": model,\n",
    "        \"scaler\": scaler,\n",
    "        \"train\": train,\n",
    "        \"trainX\": trainX,\n",
    "        \"trainY\": trainY,\n",
    "        \"trainPredict\": trainPredict,\n",
    "        \"trainScore\": trainScore,\n",
    "        \"test\": test,\n",
    "        \"testX\": testX,\n",
    "        \"testY\": testY,\n",
    "        \"testPredict\": testPredict,\n",
    "        \"testScore\": testScore\n",
    "    }\n",
    "\n",
    "def plot_lstm(model, scaler, train, test, look_back=1.0, title=\"Cases\"):\n",
    "    data = np.concatenate((train, test))\n",
    "    \n",
    "    # scale\n",
    "    train = scaler.transform(train)\n",
    "    test = scaler.transform(test) if (len(test) > 0) else []\n",
    "    \n",
    "    # prep data\n",
    "    trainX, _ = create_dataset_for_lstm(train, look_back=look_back)\n",
    "    testX, _ = (create_dataset_for_lstm(test, look_back=look_back)) if (len(test) > 0) else ([], [])\n",
    "    \n",
    "    # make predictions\n",
    "    trainPredict = model.predict(trainX).clip(0)\n",
    "    testPredict = model.predict(testX).clip(0) if (len(testX) > 0) else []\n",
    "    \n",
    "    # invert predictions\n",
    "    trainPredict = scaler.inverse_transform(trainPredict)\n",
    "    testPredict = scaler.inverse_transform(testPredict) if (len(testPredict) > 0) else []\n",
    "    \n",
    "    # plot baseline and predictions\n",
    "    plt.plot(range(1, len(data) + 1), data, label=\"Actual\")\n",
    "    plt.plot(range(1 + look_back, len(trainPredict) + look_back + 1), trainPredict, label=\"Pred Train\")\n",
    "    plt.plot(range(1 + 2*look_back + len(trainPredict) + 1, len(data)), testPredict, label=\"Pred Test\")\n",
    "    plt.title(\"Fit of LSTM Preds to Actual \" + title)\n",
    "    plt.ylabel(title)\n",
    "    plt.xlabel(\"Days\")\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 313/313 [1:12:25<00:00, 13.88s/it]\n"
     ]
    }
   ],
   "source": [
    "# settings\n",
    "train_split_factor = 0.8\n",
    "look_back = 1\n",
    "verbose = 0\n",
    "epochs = 50\n",
    "\n",
    "# entities\n",
    "geo_ids = train[\"geo_id\"].unique()\n",
    "\n",
    "# train\n",
    "results_dict = {}\n",
    "for geo_id in tqdm(geo_ids):\n",
    "    # get train and test data\n",
    "    train_size, test_size, case_train, case_test, fat_train, fat_test = get_entity_data_split(geo_id, train_split_factor=train_split_factor)\n",
    "    \n",
    "    if (verbose): print(\"train case\")\n",
    "    case_results = train_lstm(case_train, test=case_test, look_back=look_back, epochs=epochs, verbose=verbose)\n",
    "    if (verbose): print(\"train fat\")\n",
    "    fat_results = train_lstm(fat_train, test=fat_test, look_back=look_back, epochs=epochs, verbose=verbose)\n",
    "    \n",
    "    results_dict[geo_id] = (case_results, fat_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_result(geo_id, results_dict=results_dict, features=(\"case\", \"fat\")):\n",
    "    # get train and test data\n",
    "    train_size, test_size, case_train, case_test, fat_train, fat_test = get_entity_data_split(geo_id, train_split_factor=train_split_factor)\n",
    "\n",
    "    # get results\n",
    "    case_results, fat_results = results_dict[geo_id]\n",
    "\n",
    "    # print and plot\n",
    "    print(geo_id)\n",
    "    # case\n",
    "    if (\"case\" in features):\n",
    "        print(\"Confirmed Cases:\")\n",
    "        print(\"  trainScore: \", case_results[\"trainScore\"])\n",
    "        print(\"  testScore: \", case_results[\"testScore\"])\n",
    "        plot_lstm(case_results[\"model\"], case_results[\"scaler\"], case_results[\"train\"], case_results[\"test\"], look_back=look_back, title=\"Confirmed Cases\")\n",
    "    # fat\n",
    "    if (\"fat\" in features):\n",
    "        print(\"Fatalities:\")\n",
    "        print(\"  trainScore: \", fat_results[\"trainScore\"])\n",
    "        print(\"  testScore: \", fat_results[\"testScore\"])\n",
    "        plot_lstm(fat_results[\"model\"], fat_results[\"scaler\"], fat_results[\"train\"], fat_results[\"test\"], look_back=look_back, title=\"Fatalities\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of entity and results\n",
    "# entity: (geo_id, (case_results, fat_results))\n",
    "#   result: (model, scaler, trainPredict, testPredict, trainScore, testScore)\n",
    "results_list = list(results_dict.items())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Chosen Examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plot_result(\"China_Hubei\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plot_result(\"Spain_nan\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plot_result(\"Germany_nan\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confirmed Cases: 3 Best Predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort results by case_model testScore asc\n",
    "results_list.sort(key=lambda entity: entity[1][0][\"testScore\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "China_Shandong\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.34776143471783644\n",
      "  testScore:  0.0010759425697897547\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "China_Tibet\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.08410617868753063\n",
      "  testScore:  0.001485584261727513\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "China_Hebei\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.39402090493173536\n",
      "  testScore:  0.0016827216223320029\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(0, 3): plot_result(results_list[i][0], features=(\"case\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confirmed Cases: 3 Worst Predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort results by case_model testScore desc\n",
    "results_list.sort(key=lambda entity: entity[1][0][\"testScore\"], reverse=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Burma_nan\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.27815833509852084\n",
      "  testScore:  2.7556524612111906\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Guinea_nan\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.13314460174432868\n",
      "  testScore:  2.566327043311944\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Niger_nan\n",
      "Confirmed Cases:\n",
      "  trainScore:  0.12399721610194567\n",
      "  testScore:  2.2197210205667846\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(0, 3): plot_result(results_list[i][0], features=(\"case\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fatalities: 3 Best Predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort results by fat_model testScore asc\n",
    "results_list.sort(key=lambda entity: entity[1][1][\"testScore\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Guinea_nan\n",
      "Fatalities:\n",
      "  trainScore:  0.0\n",
      "  testScore:  0.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sierra Leone_nan\n",
      "Fatalities:\n",
      "  trainScore:  0.0\n",
      "  testScore:  0.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Guinea-Bissau_nan\n",
      "Fatalities:\n",
      "  trainScore:  0.0\n",
      "  testScore:  0.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(0, 3): plot_result(results_list[i][0], features=(\"fat\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fatalities: 3 Worst Predicted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort results by fat_model testScore desc\n",
    "results_list.sort(key=lambda entity: entity[1][1][\"testScore\"], reverse=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "US_Rhode Island\n",
      "Fatalities:\n",
      "  trainScore:  0.0\n",
      "  testScore:  2.9748976743220648\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "South Africa_nan\n",
      "Fatalities:\n",
      "  trainScore:  0.08547956649683079\n",
      "  testScore:  2.5816829736642157\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "US_Maine\n",
      "Fatalities:\n",
      "  trainScore:  0.08540448773659584\n",
      "  testScore:  2.480920121803671\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(0, 3): plot_result(results_list[i][0], features=(\"fat\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Overall Model Performance\n",
    "Model Performance over all predictions form all countries/provinces"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "allTrainY = []\n",
    "allTrainPredict = []\n",
    "\n",
    "allTestY = []\n",
    "allTestPredict = []\n",
    "\n",
    "# cumulate Y_true and Y_pred for train and test\n",
    "for geo_id, results in results_dict.items():\n",
    "    \n",
    "    for result in results: # case and fat\n",
    "        # train\n",
    "        trainY = result[\"trainY\"]\n",
    "        trainPredict = result[\"trainPredict\"]\n",
    "        if (trainY.size == trainPredict.size):\n",
    "            allTrainY.extend(trainY)\n",
    "            allTrainPredict.extend(trainPredict)\n",
    "        else:\n",
    "            print(\"Warning: trainY.size != trainPredict.size: \", trainY.size, \"!=\", trainPredict.size)\n",
    "\n",
    "        # test\n",
    "        testY = result[\"testY\"]\n",
    "        testPredict = result[\"testPredict\"]\n",
    "        if (testY.size == testPredict.size):\n",
    "            allTestY.extend(testY)\n",
    "            allTestPredict.extend(testPredict)\n",
    "        else:\n",
    "            print(\"Warning: testY.size != testPredict.size: \", testY.size, \"!=\", testPredict.size)\n",
    "\n",
    "# make nparrays\n",
    "allTrainY = np.array(allTrainY)\n",
    "allTrainPredict = np.array(allTrainPredict)\n",
    "allTestY = np.array(allTestY)\n",
    "allTestPredict = np.array(allTestPredict)\n",
    "# reshape for calculate_loss()\n",
    "allTrainY = allTrainY.reshape((1, allTrainY.size))\n",
    "allTrainPredict = allTrainPredict.reshape((allTrainPredict.size, 1))\n",
    "allTestY = allTestY.reshape((1, allTestY.size))\n",
    "allTestPredict = allTestPredict.reshape((allTestPredict.size, 1))\n",
    "\n",
    "# calculate overall score (loss) for train and test\n",
    "overallTrainScore = calculate_loss(allTrainY, allTrainPredict)\n",
    "overallTestScore = calculate_loss(allTestY, allTestPredict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "overallTrainScore:  0.7560530920065105\n",
      "overallTestScore:  0.773211889891129\n"
     ]
    }
   ],
   "source": [
    "print(\"overallTrainScore: \", overallTrainScore)\n",
    "print(\"overallTestScore: \", overallTestScore)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}