Region3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "wmXZjlD1PTzt"
   },
   "source": [
    "Dengue AI Project\n",
    "---\n",
    "Philippines: Region 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3080,
     "status": "ok",
     "timestamp": 1600944488243,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "_e6funNkn-69"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "k3aP2wI6xjd9"
   },
   "source": [
    "Import dataset.\n",
    "---\n",
    "*   Independent variables = Humidity, Rainfall, Temperature.\n",
    "*   Dependent variable = Dengue count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3057,
     "status": "ok",
     "timestamp": 1600944488245,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "RA4etPhfxc2m"
   },
   "outputs": [],
   "source": [
    "Philippines = pd.read_csv('dengue_climate.csv')\n",
    "Region = Philippines['Region']\n",
    "Year = Philippines['Year']\n",
    "Month = Philippines['Month']\n",
    "climate = ['Humidity', 'Rainfall', 'Temperature']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "CB__N0Y5NjuW"
   },
   "source": [
    "\n",
    "Log transformation\n",
    "---\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3027,
     "status": "ok",
     "timestamp": 1600944488246,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "yOWnkSOd41oL"
   },
   "outputs": [],
   "source": [
    "Humidity = np.log(Philippines.Humidity +1)\n",
    "Rainfall = np.log(Philippines.Rainfall +1)\n",
    "Temperature = np.log(Philippines.Temperature +1)\n",
    "Dengue_Count = np.log(Philippines.Dengue_Count +1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3005,
     "status": "ok",
     "timestamp": 1600944488247,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "4GOPJAGwCCCP"
   },
   "outputs": [],
   "source": [
    "Humidity = pd.DataFrame(Humidity, columns=['Humidity'])\n",
    "Rainfall = pd.DataFrame(Rainfall, columns=['Rainfall'])\n",
    "Temperature = pd.DataFrame(Temperature, columns=['Temperature'])\n",
    "Dengue_Count = pd.DataFrame(Dengue_Count, columns=['Dengue_Count'])\n",
    "Philippines = Humidity.join(Rainfall.join(Temperature.join(Dengue_Count)))\n",
    "Philippines = pd.concat([Region, Year, Month, Philippines], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jyoBqxDePsn1"
   },
   "source": [
    "Create training and test set\n",
    "---\n",
    "* Training set = Year 2013 to 2017\n",
    "* Test set = Year 2018\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2986,
     "status": "ok",
     "timestamp": 1600944488248,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "Gcr7FkxYqTHi"
   },
   "outputs": [],
   "source": [
    "R3_trainset = Philippines[(Philippines['Year'] < 2018) & (Philippines['Region'] == 'Region03')]\n",
    "R3_train_climate = R3_trainset.loc[:, climate].values\n",
    "R3_train_dengue = R3_trainset.iloc[:, -1].values\n",
    "R3_testset = Philippines[(Philippines['Year'] == 2018) & (Philippines['Region'] == 'Region03')]\n",
    "R3_test_climate = R3_testset.loc[:, climate].values\n",
    "R3_test_dengue = R3_testset.iloc[:, -1].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "BAuXmjzexi5k"
   },
   "source": [
    "Create artificial neural network.\n",
    "---\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5583,
     "status": "ok",
     "timestamp": 1600944490862,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "ZwGcFZekyB6h",
    "outputId": "71b7bc99-4a03-4dc6-877c-d8016e62afb8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60 samples\n",
      "Epoch 1/500\n",
      "60/60 [==============================] - 1s 16ms/sample - loss: 2.4799\n",
      "Epoch 2/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 2.3905\n",
      "Epoch 3/500\n",
      "60/60 [==============================] - 0s 165us/sample - loss: 2.3030\n",
      "Epoch 4/500\n",
      "60/60 [==============================] - 0s 111us/sample - loss: 2.2179\n",
      "Epoch 5/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 2.1349\n",
      "Epoch 6/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 2.0542\n",
      "Epoch 7/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 1.9755\n",
      "Epoch 8/500\n",
      "60/60 [==============================] - 0s 132us/sample - loss: 1.8993\n",
      "Epoch 9/500\n",
      "60/60 [==============================] - 0s 128us/sample - loss: 1.8254\n",
      "Epoch 10/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 1.7533\n",
      "Epoch 11/500\n",
      "60/60 [==============================] - 0s 75us/sample - loss: 1.6837\n",
      "Epoch 12/500\n",
      "60/60 [==============================] - 0s 75us/sample - loss: 1.6161\n",
      "Epoch 13/500\n",
      "60/60 [==============================] - 0s 78us/sample - loss: 1.5506\n",
      "Epoch 14/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 1.4873\n",
      "Epoch 15/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 1.4260\n",
      "Epoch 16/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 1.3665\n",
      "Epoch 17/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 1.3089\n",
      "Epoch 18/500\n",
      "60/60 [==============================] - 0s 78us/sample - loss: 1.2536\n",
      "Epoch 19/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 1.1994\n",
      "Epoch 20/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 1.1476\n",
      "Epoch 21/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 1.0976\n",
      "Epoch 22/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 1.0488\n",
      "Epoch 23/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 1.0018\n",
      "Epoch 24/500\n",
      "60/60 [==============================] - 0s 76us/sample - loss: 0.9567\n",
      "Epoch 25/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.9129\n",
      "Epoch 26/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.8708\n",
      "Epoch 27/500\n",
      "60/60 [==============================] - 0s 74us/sample - loss: 0.8302\n",
      "Epoch 28/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.7910\n",
      "Epoch 29/500\n",
      "60/60 [==============================] - 0s 119us/sample - loss: 0.7531\n",
      "Epoch 30/500\n",
      "60/60 [==============================] - 0s 108us/sample - loss: 0.7170\n",
      "Epoch 31/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.6821\n",
      "Epoch 32/500\n",
      "60/60 [==============================] - 0s 117us/sample - loss: 0.6483\n",
      "Epoch 33/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.6161\n",
      "Epoch 34/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.5852\n",
      "Epoch 35/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.5552\n",
      "Epoch 36/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.5268\n",
      "Epoch 37/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.4995\n",
      "Epoch 38/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.4733\n",
      "Epoch 39/500\n",
      "60/60 [==============================] - 0s 76us/sample - loss: 0.4481\n",
      "Epoch 40/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.4241\n",
      "Epoch 41/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.4013\n",
      "Epoch 42/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 0.3795\n",
      "Epoch 43/500\n",
      "60/60 [==============================] - 0s 76us/sample - loss: 0.3585\n",
      "Epoch 44/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.3387\n",
      "Epoch 45/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.3197\n",
      "Epoch 46/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.3016\n",
      "Epoch 47/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.2843\n",
      "Epoch 48/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.2679\n",
      "Epoch 49/500\n",
      "60/60 [==============================] - 0s 75us/sample - loss: 0.2524\n",
      "Epoch 50/500\n",
      "60/60 [==============================] - 0s 78us/sample - loss: 0.2375\n",
      "Epoch 51/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.2233\n",
      "Epoch 52/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 0.2100\n",
      "Epoch 53/500\n",
      "60/60 [==============================] - 0s 78us/sample - loss: 0.1975\n",
      "Epoch 54/500\n",
      "60/60 [==============================] - 0s 74us/sample - loss: 0.1853\n",
      "Epoch 55/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.1742\n",
      "Epoch 56/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.1635\n",
      "Epoch 57/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.1534\n",
      "Epoch 58/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.1440\n",
      "Epoch 59/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.1350\n",
      "Epoch 60/500\n",
      "60/60 [==============================] - 0s 74us/sample - loss: 0.1266\n",
      "Epoch 61/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.1187\n",
      "Epoch 62/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.1114\n",
      "Epoch 63/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.1042\n",
      "Epoch 64/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.101 - 0s 77us/sample - loss: 0.0978\n",
      "Epoch 65/500\n",
      "60/60 [==============================] - 0s 134us/sample - loss: 0.0916\n",
      "Epoch 66/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0858\n",
      "Epoch 67/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0805\n",
      "Epoch 68/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0753\n",
      "Epoch 69/500\n",
      "60/60 [==============================] - 0s 144us/sample - loss: 0.0706\n",
      "Epoch 70/500\n",
      "60/60 [==============================] - 0s 181us/sample - loss: 0.0661\n",
      "Epoch 71/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.0621\n",
      "Epoch 72/500\n",
      "60/60 [==============================] - 0s 78us/sample - loss: 0.0581\n",
      "Epoch 73/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 0.0545\n",
      "Epoch 74/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0512\n",
      "Epoch 75/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0481\n",
      "Epoch 76/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0452\n",
      "Epoch 77/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0426\n",
      "Epoch 78/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0401\n",
      "Epoch 79/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0379\n",
      "Epoch 80/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0358\n",
      "Epoch 81/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0338\n",
      "Epoch 82/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0320\n",
      "Epoch 83/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.0303\n",
      "Epoch 84/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0288\n",
      "Epoch 85/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0273\n",
      "Epoch 86/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0260\n",
      "Epoch 87/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0248\n",
      "Epoch 88/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0236\n",
      "Epoch 89/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0227\n",
      "Epoch 90/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0217\n",
      "Epoch 91/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.022 - 0s 95us/sample - loss: 0.0209\n",
      "Epoch 92/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0201\n",
      "Epoch 93/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0194\n",
      "Epoch 94/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0187\n",
      "Epoch 95/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0182\n",
      "Epoch 96/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0176\n",
      "Epoch 97/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0171\n",
      "Epoch 98/500\n",
      "60/60 [==============================] - 0s 74us/sample - loss: 0.0166\n",
      "Epoch 99/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0162\n",
      "Epoch 100/500\n",
      "60/60 [==============================] - 0s 122us/sample - loss: 0.0159\n",
      "Epoch 101/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0155\n",
      "Epoch 102/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0152\n",
      "Epoch 103/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0149\n",
      "Epoch 104/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0147\n",
      "Epoch 105/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0145\n",
      "Epoch 106/500\n",
      "60/60 [==============================] - 0s 184us/sample - loss: 0.0142\n",
      "Epoch 107/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0141\n",
      "Epoch 108/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0139\n",
      "Epoch 109/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0138\n",
      "Epoch 110/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0136\n",
      "Epoch 111/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0135\n",
      "Epoch 112/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0134\n",
      "Epoch 113/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0133\n",
      "Epoch 114/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0132\n",
      "Epoch 115/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0131\n",
      "Epoch 116/500\n",
      "60/60 [==============================] - 0s 134us/sample - loss: 0.0130\n",
      "Epoch 117/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0129\n",
      "Epoch 118/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0129\n",
      "Epoch 119/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0128\n",
      "Epoch 120/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0128\n",
      "Epoch 121/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0127\n",
      "Epoch 122/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0127\n",
      "Epoch 123/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0127\n",
      "Epoch 124/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0126\n",
      "Epoch 125/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0126\n",
      "Epoch 126/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0126\n",
      "Epoch 127/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0126\n",
      "Epoch 128/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.012 - 0s 87us/sample - loss: 0.0126\n",
      "Epoch 129/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0125\n",
      "Epoch 130/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0125\n",
      "Epoch 131/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0125\n",
      "Epoch 132/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0125\n",
      "Epoch 133/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0125\n",
      "Epoch 134/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0125\n",
      "Epoch 135/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0125\n",
      "Epoch 136/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0125\n",
      "Epoch 137/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0125\n",
      "Epoch 138/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 139/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 140/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n",
      "Epoch 141/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 142/500\n",
      "60/60 [==============================] - 0s 107us/sample - loss: 0.0124\n",
      "Epoch 143/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 144/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 145/500\n",
      "60/60 [==============================] - 0s 75us/sample - loss: 0.0124\n",
      "Epoch 146/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 82us/sample - loss: 0.0124\n",
      "Epoch 147/500\n",
      "60/60 [==============================] - 0s 73us/sample - loss: 0.0124\n",
      "Epoch 148/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 149/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 150/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0124\n",
      "Epoch 151/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 0.0124\n",
      "Epoch 152/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n",
      "Epoch 153/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 154/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 155/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 156/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 157/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 158/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 159/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 160/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 161/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 162/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 163/500\n",
      "60/60 [==============================] - 0s 151us/sample - loss: 0.0124\n",
      "Epoch 164/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 165/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 166/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0124\n",
      "Epoch 167/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 168/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0124\n",
      "Epoch 169/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 170/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 171/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0124\n",
      "Epoch 172/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 173/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 174/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 175/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 176/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 177/500\n",
      "60/60 [==============================] - 0s 193us/sample - loss: 0.0124\n",
      "Epoch 178/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 179/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n",
      "Epoch 180/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 181/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0124\n",
      "Epoch 182/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.0124\n",
      "Epoch 183/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 184/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 185/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0124\n",
      "Epoch 186/500\n",
      "60/60 [==============================] - 0s 77us/sample - loss: 0.0124\n",
      "Epoch 187/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 188/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 189/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 190/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 191/500\n",
      "60/60 [==============================] - 0s 191us/sample - loss: 0.0124\n",
      "Epoch 192/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 193/500\n",
      "60/60 [==============================] - 0s 83us/sample - loss: 0.0124\n",
      "Epoch 194/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 195/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 196/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.012 - 0s 117us/sample - loss: 0.0124\n",
      "Epoch 197/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 198/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 199/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 200/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n",
      "Epoch 201/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 202/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 203/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 204/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 205/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 206/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 207/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 208/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 209/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 210/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 211/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 212/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 213/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 214/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 215/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 216/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 217/500\n",
      "60/60 [==============================] - 0s 232us/sample - loss: 0.0124\n",
      "Epoch 218/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 219/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 220/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 221/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n",
      "Epoch 222/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 223/500\n",
      "60/60 [==============================] - 0s 87us/sample - loss: 0.0124\n",
      "Epoch 224/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 225/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 226/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 227/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 228/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 229/500\n",
      "60/60 [==============================] - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 230/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 231/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 232/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 233/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 234/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 235/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 236/500\n",
      "60/60 [==============================] - 0s 114us/sample - loss: 0.0124\n",
      "Epoch 237/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 238/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 239/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 240/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 241/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 242/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 243/500\n",
      "60/60 [==============================] - 0s 189us/sample - loss: 0.0124\n",
      "Epoch 244/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 245/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 246/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 247/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 248/500\n",
      "60/60 [==============================] - 0s 131us/sample - loss: 0.0124\n",
      "Epoch 249/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 250/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 251/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 252/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 253/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 254/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 255/500\n",
      "60/60 [==============================] - 0s 204us/sample - loss: 0.0124\n",
      "Epoch 256/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 257/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 258/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 259/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 260/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 261/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 262/500\n",
      "60/60 [==============================] - 0s 108us/sample - loss: 0.0124\n",
      "Epoch 263/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 264/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 88us/sample - loss: 0.0124\n",
      "Epoch 265/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 266/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 267/500\n",
      "60/60 [==============================] - 0s 202us/sample - loss: 0.0124\n",
      "Epoch 268/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 130us/sample - loss: 0.0124\n",
      "Epoch 269/500\n",
      "60/60 [==============================] - 0s 130us/sample - loss: 0.0124\n",
      "Epoch 270/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 271/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 272/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 273/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 274/500\n",
      "60/60 [==============================] - 0s 113us/sample - loss: 0.0124\n",
      "Epoch 275/500\n",
      "60/60 [==============================] - 0s 111us/sample - loss: 0.0124\n",
      "Epoch 276/500\n",
      "60/60 [==============================] - 0s 114us/sample - loss: 0.0124\n",
      "Epoch 277/500\n",
      "60/60 [==============================] - 0s 117us/sample - loss: 0.0124\n",
      "Epoch 278/500\n",
      "60/60 [==============================] - 0s 134us/sample - loss: 0.0124\n",
      "Epoch 279/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 280/500\n",
      "60/60 [==============================] - 0s 115us/sample - loss: 0.0124\n",
      "Epoch 281/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 282/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 283/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 284/500\n",
      "60/60 [==============================] - 0s 117us/sample - loss: 0.0124\n",
      "Epoch 285/500\n",
      "60/60 [==============================] - 0s 116us/sample - loss: 0.0124\n",
      "Epoch 286/500\n",
      "60/60 [==============================] - 0s 116us/sample - loss: 0.0124\n",
      "Epoch 287/500\n",
      "60/60 [==============================] - 0s 129us/sample - loss: 0.0124\n",
      "Epoch 288/500\n",
      "60/60 [==============================] - 0s 148us/sample - loss: 0.0124\n",
      "Epoch 289/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 290/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 291/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 292/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 293/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 294/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.010 - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 295/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 296/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.013 - 0s 172us/sample - loss: 0.0124\n",
      "Epoch 297/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 298/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 299/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 300/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 301/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 302/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.0124\n",
      "Epoch 303/500\n",
      "60/60 [==============================] - 0s 81us/sample - loss: 0.0124\n",
      "Epoch 304/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 305/500\n",
      "60/60 [==============================] - 0s 141us/sample - loss: 0.0124\n",
      "Epoch 306/500\n",
      "60/60 [==============================] - 0s 111us/sample - loss: 0.0124\n",
      "Epoch 307/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 308/500\n",
      "60/60 [==============================] - 0s 147us/sample - loss: 0.0124\n",
      "Epoch 309/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 310/500\n",
      "60/60 [==============================] - 0s 74us/sample - loss: 0.0124\n",
      "Epoch 311/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 312/500\n",
      "60/60 [==============================] - 0s 75us/sample - loss: 0.0124\n",
      "Epoch 313/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 314/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 315/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 316/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 317/500\n",
      "60/60 [==============================] - 0s 145us/sample - loss: 0.0124\n",
      "Epoch 318/500\n",
      "60/60 [==============================] - 0s 113us/sample - loss: 0.0124\n",
      "Epoch 319/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 320/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 321/500\n",
      "60/60 [==============================] - 0s 132us/sample - loss: 0.0124\n",
      "Epoch 322/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 323/500\n",
      "60/60 [==============================] - 0s 148us/sample - loss: 0.0124\n",
      "Epoch 324/500\n",
      "60/60 [==============================] - 0s 121us/sample - loss: 0.0124\n",
      "Epoch 325/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 326/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 327/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 328/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 329/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 330/500\n",
      "60/60 [==============================] - 0s 164us/sample - loss: 0.0124\n",
      "Epoch 331/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 332/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 333/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 334/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 335/500\n",
      "60/60 [==============================] - 0s 113us/sample - loss: 0.0124\n",
      "Epoch 336/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 337/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 338/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 339/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 340/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 341/500\n",
      "60/60 [==============================] - 0s 143us/sample - loss: 0.0124\n",
      "Epoch 342/500\n",
      "60/60 [==============================] - 0s 232us/sample - loss: 0.0124\n",
      "Epoch 343/500\n",
      "60/60 [==============================] - 0s 166us/sample - loss: 0.0124\n",
      "Epoch 344/500\n",
      "60/60 [==============================] - 0s 223us/sample - loss: 0.0124\n",
      "Epoch 345/500\n",
      "60/60 [==============================] - 0s 131us/sample - loss: 0.0124\n",
      "Epoch 346/500\n",
      "60/60 [==============================] - 0s 226us/sample - loss: 0.0124\n",
      "Epoch 347/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 0.0124\n",
      "Epoch 348/500\n",
      "60/60 [==============================] - 0s 126us/sample - loss: 0.0124\n",
      "Epoch 349/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 350/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 351/500\n",
      "60/60 [==============================] - 0s 115us/sample - loss: 0.0124\n",
      "Epoch 352/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 353/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 354/500\n",
      "60/60 [==============================] - 0s 108us/sample - loss: 0.0124\n",
      "Epoch 355/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.012 - 0s 206us/sample - loss: 0.0124\n",
      "Epoch 356/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 357/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 358/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 121us/sample - loss: 0.0124\n",
      "Epoch 359/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 360/500\n",
      "60/60 [==============================] - 0s 217us/sample - loss: 0.0124\n",
      "Epoch 361/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 362/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 363/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 364/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 365/500\n",
      "60/60 [==============================] - 0s 139us/sample - loss: 0.0124\n",
      "Epoch 366/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 367/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 368/500\n",
      "60/60 [==============================] - 0s 85us/sample - loss: 0.0124\n",
      "Epoch 369/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 370/500\n",
      "60/60 [==============================] - 0s 133us/sample - loss: 0.0124\n",
      "Epoch 371/500\n",
      "60/60 [==============================] - 0s 279us/sample - loss: 0.0124\n",
      "Epoch 372/500\n",
      "60/60 [==============================] - 0s 111us/sample - loss: 0.0124\n",
      "Epoch 373/500\n",
      "60/60 [==============================] - 0s 114us/sample - loss: 0.0124\n",
      "Epoch 374/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 375/500\n",
      "60/60 [==============================] - 0s 124us/sample - loss: 0.0124\n",
      "Epoch 376/500\n",
      "60/60 [==============================] - 0s 138us/sample - loss: 0.0124\n",
      "Epoch 377/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 0.0124\n",
      "Epoch 378/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 379/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 0.0124\n",
      "Epoch 380/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 0.0124\n",
      "Epoch 381/500\n",
      "60/60 [==============================] - 0s 302us/sample - loss: 0.0124\n",
      "Epoch 382/500\n",
      "60/60 [==============================] - 0s 177us/sample - loss: 0.0124\n",
      "Epoch 383/500\n",
      "60/60 [==============================] - 0s 112us/sample - loss: 0.0124\n",
      "Epoch 384/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 385/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 386/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 387/500\n",
      "60/60 [==============================] - 0s 135us/sample - loss: 0.0124\n",
      "Epoch 388/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 389/500\n",
      "60/60 [==============================] - 0s 125us/sample - loss: 0.0124\n",
      "Epoch 390/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 391/500\n",
      "60/60 [==============================] - 0s 236us/sample - loss: 0.0124\n",
      "Epoch 392/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.012 - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 393/500\n",
      "60/60 [==============================] - 0s 90us/sample - loss: 0.0124\n",
      "Epoch 394/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 395/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 396/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 397/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 398/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 399/500\n",
      "60/60 [==============================] - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 400/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 401/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 402/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 403/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 404/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.011 - 0s 92us/sample - loss: 0.0124\n",
      "Epoch 405/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 406/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 407/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 408/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 409/500\n",
      "60/60 [==============================] - 0s 117us/sample - loss: 0.0124\n",
      "Epoch 410/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 411/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 412/500\n",
      "60/60 [==============================] - 0s 108us/sample - loss: 0.0124\n",
      "Epoch 413/500\n",
      "60/60 [==============================] - 0s 259us/sample - loss: 0.0124\n",
      "Epoch 414/500\n",
      "60/60 [==============================] - 0s 123us/sample - loss: 0.0124\n",
      "Epoch 415/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 416/500\n",
      "60/60 [==============================] - 0s 120us/sample - loss: 0.0124\n",
      "Epoch 417/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 418/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 419/500\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.010 - 0s 144us/sample - loss: 0.0124\n",
      "Epoch 420/500\n",
      "60/60 [==============================] - 0s 143us/sample - loss: 0.0124\n",
      "Epoch 421/500\n",
      "60/60 [==============================] - 0s 151us/sample - loss: 0.0124\n",
      "Epoch 422/500\n",
      "60/60 [==============================] - 0s 128us/sample - loss: 0.0124\n",
      "Epoch 423/500\n",
      "60/60 [==============================] - 0s 193us/sample - loss: 0.0124\n",
      "Epoch 424/500\n",
      "60/60 [==============================] - 0s 127us/sample - loss: 0.0124\n",
      "Epoch 425/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 426/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 427/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 428/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 429/500\n",
      "60/60 [==============================] - 0s 119us/sample - loss: 0.0124\n",
      "Epoch 430/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 431/500\n",
      "60/60 [==============================] - 0s 107us/sample - loss: 0.0124\n",
      "Epoch 432/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 433/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 434/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 435/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 436/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 437/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 438/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 439/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 440/500\n",
      "60/60 [==============================] - 0s 101us/sample - loss: 0.0124\n",
      "Epoch 441/500\n",
      "60/60 [==============================] - 0s 102us/sample - loss: 0.0124\n",
      "Epoch 442/500\n",
      "60/60 [==============================] - 0s 104us/sample - loss: 0.0124\n",
      "Epoch 443/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 444/500\n",
      "60/60 [==============================] - 0s 100us/sample - loss: 0.0124\n",
      "Epoch 445/500\n",
      "60/60 [==============================] - 0s 110us/sample - loss: 0.0124\n",
      "Epoch 446/500\n",
      "60/60 [==============================] - 0s 114us/sample - loss: 0.0124\n",
      "Epoch 447/500\n",
      "60/60 [==============================] - 0s 116us/sample - loss: 0.0124\n",
      "Epoch 448/500\n",
      "60/60 [==============================] - 0s 80us/sample - loss: 0.0124\n",
      "Epoch 449/500\n",
      "60/60 [==============================] - 0s 122us/sample - loss: 0.0124\n",
      "Epoch 450/500\n",
      "60/60 [==============================] - 0s 144us/sample - loss: 0.0124\n",
      "Epoch 451/500\n",
      "60/60 [==============================] - 0s 91us/sample - loss: 0.0124\n",
      "Epoch 452/500\n",
      "60/60 [==============================] - 0s 140us/sample - loss: 0.0124\n",
      "Epoch 453/500\n",
      "60/60 [==============================] - 0s 125us/sample - loss: 0.0124\n",
      "Epoch 454/500\n",
      "60/60 [==============================] - 0s 105us/sample - loss: 0.0124\n",
      "Epoch 455/500\n",
      "60/60 [==============================] - 0s 164us/sample - loss: 0.0124\n",
      "Epoch 456/500\n",
      "60/60 [==============================] - 0s 135us/sample - loss: 0.0124\n",
      "Epoch 457/500\n",
      "60/60 [==============================] - 0s 120us/sample - loss: 0.0124\n",
      "Epoch 458/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 459/500\n",
      "60/60 [==============================] - 0s 118us/sample - loss: 0.0124\n",
      "Epoch 460/500\n",
      "60/60 [==============================] - 0s 111us/sample - loss: 0.0124\n",
      "Epoch 461/500\n",
      "60/60 [==============================] - 0s 133us/sample - loss: 0.0124\n",
      "Epoch 462/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 463/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 464/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 465/500\n",
      "60/60 [==============================] - 0s 124us/sample - loss: 0.0124\n",
      "Epoch 466/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 467/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 468/500\n",
      "60/60 [==============================] - 0s 98us/sample - loss: 0.0124\n",
      "Epoch 469/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 470/500\n",
      "60/60 [==============================] - 0s 89us/sample - loss: 0.0124\n",
      "Epoch 471/500\n",
      "60/60 [==============================] - 0s 94us/sample - loss: 0.0124\n",
      "Epoch 472/500\n",
      "60/60 [==============================] - 0s 86us/sample - loss: 0.0124\n",
      "Epoch 473/500\n",
      "60/60 [==============================] - 0s 108us/sample - loss: 0.0124\n",
      "Epoch 474/500\n",
      "60/60 [==============================] - 0s 82us/sample - loss: 0.0124\n",
      "Epoch 475/500\n",
      "60/60 [==============================] - 0s 99us/sample - loss: 0.0124\n",
      "Epoch 476/500\n",
      "60/60 [==============================] - 0s 106us/sample - loss: 0.0124\n",
      "Epoch 477/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 478/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 479/500\n",
      "60/60 [==============================] - 0s 96us/sample - loss: 0.0124\n",
      "Epoch 480/500\n",
      "60/60 [==============================] - 0s 79us/sample - loss: 0.0124\n",
      "Epoch 481/500\n",
      "60/60 [==============================] - 0s 97us/sample - loss: 0.0124\n",
      "Epoch 482/500\n",
      "60/60 [==============================] - 0s 95us/sample - loss: 0.0124\n",
      "Epoch 483/500\n",
      "60/60 [==============================] - 0s 128us/sample - loss: 0.0124\n",
      "Epoch 484/500\n",
      "60/60 [==============================] - 0s 150us/sample - loss: 0.0124\n",
      "Epoch 485/500\n",
      "60/60 [==============================] - 0s 134us/sample - loss: 0.0124\n",
      "Epoch 486/500\n",
      "60/60 [==============================] - 0s 121us/sample - loss: 0.0124\n",
      "Epoch 487/500\n",
      "60/60 [==============================] - 0s 125us/sample - loss: 0.0124\n",
      "Epoch 488/500\n",
      "60/60 [==============================] - 0s 149us/sample - loss: 0.0124\n",
      "Epoch 489/500\n",
      "60/60 [==============================] - 0s 93us/sample - loss: 0.0124\n",
      "Epoch 490/500\n",
      "60/60 [==============================] - 0s 155us/sample - loss: 0.0124\n",
      "Epoch 491/500\n",
      "60/60 [==============================] - 0s 130us/sample - loss: 0.0124\n",
      "Epoch 492/500\n",
      "60/60 [==============================] - 0s 103us/sample - loss: 0.0124\n",
      "Epoch 493/500\n",
      "60/60 [==============================] - 0s 109us/sample - loss: 0.0124\n",
      "Epoch 494/500\n",
      "60/60 [==============================] - 0s 135us/sample - loss: 0.0124\n",
      "Epoch 495/500\n",
      "60/60 [==============================] - 0s 131us/sample - loss: 0.0124\n",
      "Epoch 496/500\n",
      "60/60 [==============================] - 0s 117us/sample - loss: 0.0124\n",
      "Epoch 497/500\n",
      "60/60 [==============================] - 0s 112us/sample - loss: 0.0124\n",
      "Epoch 498/500\n",
      "60/60 [==============================] - 0s 123us/sample - loss: 0.0124\n",
      "Epoch 499/500\n",
      "60/60 [==============================] - 0s 133us/sample - loss: 0.0124\n",
      "Epoch 500/500\n",
      "60/60 [==============================] - 0s 84us/sample - loss: 0.0124\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7f1d4400fb00>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "R3ann = tf.keras.models.Sequential()\n",
    "R3ann.add(tf.keras.layers.Dense(units=3, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=6, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=8, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=8, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=6, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=3, activation='relu'))\n",
    "R3ann.add(tf.keras.layers.Dense(units=1))\n",
    "R3ann.compile(optimizer = 'adam', loss = 'mean_squared_logarithmic_error')\n",
    "R3ann.fit(R3_train_climate, R3_train_dengue, batch_size = 48, epochs = 500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "UlOgf-mwRdnp"
   },
   "source": [
    "Predict dengue using the test set\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5572,
     "status": "ok",
     "timestamp": 1600944490865,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "14vTMzTJydCz"
   },
   "outputs": [],
   "source": [
    "predict_R3_dengue = R3ann.predict(R3_test_climate)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7xHujE7z4oPv"
   },
   "source": [
    "\n",
    "Get the Mean Squared Logarithmic Error\n",
    "---\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 5565,
     "status": "ok",
     "timestamp": 1600944490866,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "Lv80dSDA4ofK"
   },
   "outputs": [],
   "source": [
    "msle = tf.keras.losses.MeanSquaredLogarithmicError()\n",
    "R3_msle = msle(R3_test_dengue, predict_R3_dengue).numpy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "bujqLI8DRqIn"
   },
   "source": [
    "Compare the predicted from the actual values of dengue incidence\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 627
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6903,
     "status": "ok",
     "timestamp": 1600944492212,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "nojXBmbg3C4U",
    "outputId": "c2f29456-36bc-4f43-c84d-f5df19e0644a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAElCAYAAADp4+XfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO2dd5gUZdLAf0VGEBVBRZRgQkwoIGblBHPAgAEF1gQqhyjm8zCinnf66ZkDSBAJIpgxoqCcmaSioIKigiBBSUre+v6oXhmX3WV2p3t6Qv2ep5+Z6fC+NRu6+q0oqorjOI6Tv1SKWwDHcRwnXlwROI7j5DmuCBzHcfIcVwSO4zh5jisCx3GcPMcVgeM4Tp7jisBxyomIPCYiN8Yth+OEhXgegZOLiMhsYFtgPbACeB3oqaor4pSrPIjI2cCtwHbAauA14DJVXRarYE7O4SsCJ5c5SVVrA/sC+wH/iFme8vI+cIiqbgHsBFQBbo9XJCcXcUXg5DyqOh94A1MIAIhIdRG5R0R+FJFfAnNPzYTj14rIPBH5WUQuEhEVkV2CY4NE5PaEc7uJyEwR+VVEXhKR7ROOqYhcIiLfisgSEXlYRCRJuX9S1UUJu9YDu1T8J+E4JeOKwMl5RGQH4DhgZsLuu4DdMOWwC9AQuCk4/1jgSqB9cKxtGWMfCfwLOBNoAPwAjCh22onA/sA+wXnHBNc2CpRDozLGP1RElgLLgdOB/ybznR2nPLiPwMlJAh9BPUCB2sA7wOmquiR4Il8B7KOqs4LzDwKGqWpTERkA/KKq/wiO7QJ8C+yqqjNFZBAwR1X7iMiTwGJVvTY4tzbwW3DubBFR4DBV/V9wfCQwWVXvKuf3aQh0C2T8JoUfjeNshK8InFzmFFXdHHui3x1TDAD1gc2AScET+RLMmVw/OL498FPCOInvi7M9tgoAIHBGL8ZWGEXMT3j/B6aYyoWqzg1kLL7acJyUcUXg5Dyq+i4wCLgn2LUIWAnsqapbBtsWgWMZYB6wQ8IQO5Yx/M9A46IPIlIL2BqYG5L4iVQBdo5gXCfPcUXg5Av/BY4SkRaqWgj0A+4TkW3ATC8ickxw7kjgfBFpLiKbAWXlDAwPzt1XRKoDdwIfq+rsVAUWkXOL/Aci0hi4A3g71XEdpziuCJy8QFUXAk8ROISB6zDn8UcisgwYCzQLzn0NeAAYV3ROcM3qEsYdiymK0dhKYmfg7GRkCpzFK8pwFu8BfCAiv2OhpF9jfgLHCRV3FjvOJhCR5sA0oLqqrotbHscJG18ROE4JiMipQa7BVsC/gZddCTi5iisCxymZi4EFwCwskevSeMVxnOhw05DjOE6e4ysCx3GcPMcVgeM4Tp7jisBJG0EBtgUiUiVhX9Vgnybs21NE3gyKuC0RkUkicnxwrK2IzCll/EEisiYIySzaPiuHfO1EZIaI/CEi44LY/dLObRKc80dwTfuEY3uJyBsisijxewXHqovIkyLyg4gsF5GpInJcsXG12He4sdj1A0RkmYjMF5ErE45VE5FRIjI7GKNtsbmvEZFpwbzfi8g1CccaFZtzRTDGVcHxG4odWykihSJSDyfrcUXgpJvfsAJwRRwX7EvkZeAtrA7/NkAvINka/P9R1doJW4tkLgpuaM9hOQF1gYnAM2VcMhyYgmUR/xMYJSJFJSrWYklpF5ZwXRWsZMURwBZAH2CkiDQpdt6WCd+hb8L+W4BdsWzmvwHXBkXyivgf0Jm/lrX482sCXYGtgGOBnmI9D1DVHxN/bsDeQCGWH4Gq3lns+L+B8cWqozrZiqr65ltaNqwAXB/g2YR9o7AbqQafiwrFbVnKGG2xgm8lHRsE3F5B2boDHyR8roWVodi9hHN3w5LLNk/YNwG4pNh5uxR9r03M/TlWEA+gSfD9q5Ry7s/A0Qmf+wIjSjhvDtB2E/M+ADxYyrGbgXGlHBPgO6Ag7r8p38LZfEXgpJsXgMNFZMsgRv8w4MWE44uxbN6nReQUEdk2rIlF5HMROaeUw3sCf5qRVPV3LHR0z1LO/U5Vlyfs+6yUczcl07aYYvmy2KEfRGSOiAwsMr8EP68GiXKmMK9gP/vi8xYd6woMLuXyw7CV2ujyzutkJq4InHSzCjP9nBVsLwX7gD+XBX8DZgP/B8wTkfdEZNckx7+6qKJosP15M1PVfVR1WCnX1QaWFtu3FNg8xXNLRUSqAkOBwao6I9i9COtd0BhoFYw5NGHeorkqPG/ALdj//8ASjh2KtfkcVcq1BcAozaK2n07ZVNn0KY4TOk9hzVwEq/nzF1R1DtATQER2BJ4IrjkoibHvUdU+FZBpBVCn2L46WEOYVM4tERGpBAwB1hB8V/izjPXE4OMvItITU4abB/MWzbUq4X3S8wZz98Se+A9T1Y3qJ2E3+tEl3eiDInxnAB3KM6eT2fiKwImDCZiJY1vMuVkqqvoT8DCwV8QyfQn86VgWKye9MyWYToJ9OwU35yJalHLuRgSmlyex73+6qq4t4/SiqKNKqvobVtgu0QGe9LzB3BcA1wPtAoVb/HhN7EZfmlnoVOBXYHyyczqZjysCJ+0E5p+TgJOD938iIluJyK0isouIVArs4xewoQJo0Xk1im1J9QEug+eBvUTkdBGpgVUp/TzBZJMo/zfAVODmYO5TsTaUowPZJBijWoKs1ROGeBRoDpykqiuLfa8DRKRZ8N23xhy641W1yBz0FNAn+DntjlUjHZRwffVgboBqiT8bETkXK5N9lKp+V8rP4VQsimtcKccLgKeK/96cLCdub7Vv+bNhT7e7lLD/z+gaLFpnMOYjWIGFQQ4HGgbH2wbjFN92wW6Ia4LrirZFCfN8CZxbhnztgRlYtNB4oEnCsceAxxI+NwnOWYmVh25f7Fhx+WYHxxoHn1cVk/Pc4Hgn4Hvgd+zp/ylgu4SxqwMDsHDaX4Ari32H2SXM3SQ49j0W2po472PFrn8D6FvKz6chsK6k36Fv2b15rSHHcZw8x01DjuM4eY4rAsdxnDzHFYHjOE6e44rAcRwnz8m6hLJ69eppkyZN4hbDcRwnq5g0adIiVa1f0rGsUwRNmjRh4sSJmz7RcRzH+RMR+aG0Y24achzHyXNcETiO4+Q5rggcx3HyHFcEjuM4eY4rAsdxnDzHFYHjOE6e44rAcRwnz3FF4Dj5xLRpMK60VgNOvpJ1CWWO41SQuXPhb3+DP/6AefOgTvFum06+4isCx8kH1q2DTp1g2TJTBMOGxS2Rk0G4InCcfOC222DCBHjySWjRAvr1i1siJ4NwReA4uc7YsXD77XDBBdC5M3TvDpMnw6RJcUvmZAiuCBwnl5k/327+zZvDAw/YvnPPhZo1fVXg/IkrAsfJVdavNyWwbBmMHAm1atn+LbaAs86CoUNhxYp4ZXQyAlcEjpOr/Otf8Pbb8NBDsOeefz3WrZspgWeeiUc2J6NwReA4uci778LNN9uK4PzzNz5+0EGwxx7wxBPpl83JOFwROE6usXAhnHMO7LILPPIIiGx8jog5jT/5BD7/PP0yOhmFKwLHySUKC6FrV1i82PwCm29e+rldukD16u40dlwROE5Occ898Prr8N//Wr5AWdStCx07wpAhlmTm5C2uCBwnV/jgA7jhBjjjDLj44uSu6dYNli6FUaOilc3JaFwROE4u8OuvVkKicWMz9ZTkFyiJww+H3XZzp3Ge44rAcbIdVYsMmjfPwkG32CL5a0VsVfD++/DVV9HJ6GQ0rggcJ9u5/3546SW4+25o3br81xcUQNWq0L9/+LI5WYErAsfJZj79FK69Fjp0gF69KjZG/fpw6qkweDCsWhWufE5W4IrAcbKVJUusVESDBjBgQPJ+gZLo1s38DM8/H558TtbgisBxshFVu3n/9JP5BerWTW28I4+Epk3daZynuCJwnGzksccs5PPOO+HAA1Mfr1IlUyzjx8O336Y+npNVuCJwnGxj6lTo3RuOPx6uuiq8cc87DypXdqdxHhKpIhCR3iLypYhME5HhIlKj2PHzRGShiEwNtouilMdxsp7ly+HMM2Hrrc25WynEf+EGDeDkk2HgQFizJrxxnYwnMkUgIg2BXkBrVd0LqAycXcKpz6jqvsHmjyKOUxqqcMklMGsWDB8O9eqFP0e3bla07qWXwh/byViiNg1VAWqKSBVgM+DniOdznNxlwABrOn/bbZYRHAVHHw2NGrnTOM+ITBGo6lzgHuBHYB6wVFXfLOHU00XkcxEZJSI7RiWP42Q106bBZZdB+/Zw/fXRzVO5Mlx4Ibz1Fnz/fXTzOBlFlKahrYAOQFNge6CWiHQudtrLQBNV3Qd4CxhcyljdRWSiiExcuHBhVCI7Tmby++/mF6hTB55+2m7WUXLBBeZ7ePLJaOdxMoYoTUPtge9VdaGqrgWeAw5OPEFVF6vq6uBjf6BVSQOp6hOq2lpVW9evXz9CkR0nA7nsMpgxw3oMb7tt9PPtsINFJA0YAOvWRT+fEztRKoIfgQNFZDMREaAdMD3xBBFpkPDx5OLHHSfvGTLEonhuvBHatUvfvN26WRG7MWPSN6cTG1H6CD4GRgGTgS+CuZ4QkdtE5OTgtF5BeOlnWITReVHJ4zhZx4wZcOmlcMQRcNNN6Z37+ONh++3daZwniKrGLUO5aN26tU6cODFuMRwnWlauhAMOsKfyqVOhYcP0y3DjjZa5PHs27OhxHNmOiExS1RLL03pmseNkIr17wxdfmGkoDiUAFj2kar4CJ6dxReA4mcYzz8Djj8N118Gxx8YnR5Mmllfw5JOwfn18cjiR44rAcTKJmTPNUXvwwdC3b9zSbKhw+sYbcUviRIgrAsfJFFavtv4CVapYCYmqVeOWyGoPbbutO41zHFcEjpMpXHMNTJ5sxeQaNYpbGqNqVatK+sor5rh2chJXBI6TCTz/PDz4oDmJTzopbmn+ykUXmY9g4MC4JXEiwhWB48TN7NlW1mH//eGuu+KWZmN22cU6mPXvD4WFcUvjRIArAseJkzVrzC+gatFC1arFLVHJdOtmRejefjtuSZwIcEXgOHEyZAh88gn062c9gzOVU0+1ZjjuNM5JXBE4TozowIHQvDl07Bi3KGVTvToUFMCLL8KCBXFL44SMKwLHiYlX7vsWef991nU+j7XrhN69LY0gY+nWDdautagmJ6dwReA4aWb1aujZE6ZeOZj1VGLpyV2YMsWSiXffHS6+GObOjVvKEth9dzjsMDNjZVmNMqdsXBE4Thr54Qe7lz768Hp61h6MHHcsW+/VgDZtrBXxJZdYlObOO8NVV1ntuYyiWzf49lt49924JXFCxBWB46QJVTjnHPj6axh/0zi2XDGHSuef9+fxBg3goYfs+Nlnw/jxZpqHDIra7NgRttzSncY5hisCx4mYdetg1SoQsVD8SZPgsJkDYautSkwea9oUBg2CDz6wjpG//Qa77QZ3350BK4SaNaFLFxg9GhYvjlkYJyxcEThOhMyfbwU8L73UPjdvDrvUXwrPPWfLgxo1Sr22aDWwdKnldF17rb0+9pilH8RGt24mwFNPxSiEEyauCBwnIt57D/bbDz76CNq2TTgwcqQtEc47L6lxmjSB1183U1GTJqZUmjeHJUtCFzk59t4bDjzQncY5hCsCxwmZwkL4z3+sKkOdOvDxxxaC/yeDBsGee0KrVuUa94gj4H//szbCp51mpnqAzz6L4X7crRtMn272KyfrcUXgOCHz889wxx1w+unw6af2AP0nX39tN8/zzjOnQTkRsXbCd99tn2fNgtatravl2LFpVAhnnQWbb+5O4xzBFYHjhMS339qNeIcdrJr0iBG2IvgLgwdD5crQuXMoczZubPkH8+fDUUdBu3bw4YehDF02tWrBueeameu339IwoRMlrggcJ0VUzYG7114bHpB33rmEB/71683BetxxsN12ocxdpYoVLv3mG7j/fpg2zfwRaakC0b27+TqGDk3DZE6UuCJwnBRYscKiKS+91HwCZZYMevttSxlO0klcHmrUgF694Lvv4KWXYJttbP9//mMrlUjYbz/zc7jTOOtxReA4FWT6dLPNDxtm7YXHjLECnaUyaBDUrQsnnhiZTLVrwzHH2PuffoJbb7UIo+7d7XPodOsGn39uzpB84623bDWUA0rQFYHjVJA5c+DXX+HNN6FPH0v+KpUlS6wL2TnnbEgQiJgddzRnco8epoN23RWuvDLksNNOncxfkE9O40WLzMdz9NH2etllZvbLYlwROE45WL3aYvrBnLOzZkH79klc+Mwz5codCIvttoMHHjAfQqdO8PTTm1BY5aVOHauHMWIELFsW4sAZiCoMH25LrJEj4eabrSDUww9bv4bff49bwgrjisBxkmT2bDj0UDjhBFMAAJttluTFgwZZHGnLlhFJVzZNmlgxu1mz7N69bp05uNeuDWHw7t3tJjh8eAiDZSg//WTlQM45B3baycLCbrkF7rnHFMGYMealnz8/bkkrhCsCx0mCV16xe/i331qZnZ13LsfFM2ZYenEFcwfCZPPN7XXMGHNw/+1vIfgO9t8f9tnHnMa5RmEhPPqoJQCOGwf33Wd5IHvtteGcHj3ghRfgq68s43r69PjkrSCuCBxnE9x0kz0MNm5sBeNOOaWcAxTlDpx7biTyVYQOHczJ/dlnFvzz6qspDCZiq4JJk+xJOVf4+mt7yu/Rw6ICpk2DK66w32VxTjrJSnOvWgUHH2z1QLIIVwSOswm23NKCYz74oJwrAdiQO3D88bDttpHIV1E6dbJ79w47mLnr3ntTGOzcc60yaS6sCtauhX/9C1q0gC++gAEDLCJgUz2lW7e2lV+DBuZIfvrp9MgbBqqaVVurVq3UcaJm/HjVV1+194WFKQz02muqoDp6dChyRcEff6j26KH66acpDtS1q+rmm6uuWBGKXLEwaZLqvvva76xjR9V588o/xq+/qrZta2P07ZviH1B4ABO1lPuqrwgcpxj33WfJYbfdZoEiKZn1Bw2y5IIIcwdSpWZN83e2bm2f+/SpoKmoe3dYvtwipLKNlSvhuuugTRtz+D73HDz7bMUywLfaykLLOneGG2/c0Os5kylNQ2Tq5isCJ0pmz7YHuZNPVl22LMXBfv1VtXp11V69QpEtHaxYodqihf0MrrtOdc2aclxcWKjavLnqgQdGJl8kjB+vuuuu9qUvvNB+b2FQWKjap4+Ne/TRqkuXhjNuBSGuFYGI9BaRL0VkmogMF5EaxY5XF5FnRGSmiHwsIk2ilMdxNkVRBOR9922IsKkwI0ZY4kGacwdSoVYtK1rXvTv8+9/ljCoqchp/9JHZ1jOdpUutSXTbtubLGTvWWshttVU444tYyvmTT8I771iz6jlzwhk7bErTEKluQEPge6Bm8HkkcF6xc3oAjwXvzwae2dS4viJwomSvvVQPPjikwdq0Ud1nn4yxEZeXYcNUa9dWbdxYdfXqJC9atEi1WjXVyy6LUrTUefFF1YYNVStVUr3qKtXff492vjfeMP9Jw4aqU6dGO1cpEKOPoApQU0SqAJsBPxc73gEYHLwfBbQTiTnQ2slbVC0/6LbbQhjsq6/gk0/g/PNjzx2oKEVRRQ8+CNWq2c9nk5UUtt7aGjEMGZIBDZZLYMECy4Tu0MHqPn30kf3Sk84MrCBHH21dhcBWBm+8Ee185SQyRaCqc4F7gB+BecBSVX2z2GkNgZ+C89cBS4GNynaJSHcRmSgiExcuXBiVyE6eI2IF29q1C2GwwYOtRvQ554QwWHzstpuFyINZOI44IglTUffuVtBo1KjI5UsaVVNOzZtbzafbboOJEy0ZLl3ss4+1q9tpJ4vXffLJ9M29CSJTBCKyFfbE3xTYHqglIhXqxqGqT6hqa1VtXb9+/TDFdBzAEkj79g2pZPO6dXbTOeGEDfWgc4A6dZJMQDviCKtwlyk5BT/8YD0gunaF3XeHKVMsmqdatfTL0rChNbNu3x4uushCtDT+6qVRmobaA9+r6kJVXQs8Bxxc7Jy5wI4AgfloC2BxhDI5TolMmGAZxBMnhjDYm2/CvHlZ5SROhjPPNFNRw4am466/vpSoSBELmZwwwcprxEVhodm19tzTzDIPPmgy7bFHfDKBadSXXzZFcMcdFma6enWsIiWtCESkloiUkFtdKj8CB4rIZoHdvx1QvAjHS0BRW++OwDuBU8Nx0srQoRYxc/LJIQw2aBDUq2fZxDnGbruZWb0oqqjI7L0RBQVQtWp8q4Lp080W36uXvX75JfTsGXLp1RSoWtVKd99xh9X6OPpoq2keF6V5kTElcQ4wBliA2fIXAF8BdwO7lHZtwhi3AjOAacAQoDpwG3BycLwG8CwwE/gE2GlTY3rUkBM2q1apbrmlaufOIQy2eLFFzVxxRQiDZTaTJ294P3t2CSd07Ki69db2A04Xq1dbNm+1aqp166o+9VTmR20NHWry7r676nffRTYNZUQNiZbyAC4i7wJjgReBaapaGOyvC/wtUBLPq2paC2q0bt1aJ4ayfncc44UXrJz8q6+aKTklHnkE/v53mDrVatXkARMnWp21K680P0vVqsGBt96yJ9327TfRui1EPv/cVgNnnWWNGLLFR/Pee1bNsGpVK3UbgRNbRCapausSj5WhCKqq2fbLGniT54SNKwInbB54wBLIvv3WAn1Sok0bM5xPmRKKbNnAypXQuzc8/rgphBEjrDsahYWmYb/+On3C1K5tzp5QbHxpZsYMexL55RfLbOzQIdThK6QIEi6uW8Lu5elWAEW4InCiYP36kqsLl4svv7Q69f/9L1x+eShyZRPDh5vvoFo1K7h6wglxS5SF/PKLxetOnAj3329tMEOiLEWQjOdkMrAQ+Ab4Nng/W0Qmi0ir0KR0nBgo6i6YshIAcxLnQO5ARSlKQNtxx9xqS5BWtt3WGuCcfLI5uq+80lZWEZOMIngLOF5V66nq1sBxwCtYeYhHohTOcaKmQwc47bQQBirKHTjpJMjjXJeiqKIbbrDPH3wAPxevJ+CUTa1a1gavVy+zWZ5xRuRZ2skoggNV9c98aLXs4INU9SMsCshxspKff7ZaYIldByvMG2/Ysj7HcgcqQo0atsJat85C5Dt2tPdOOahc2UxD991nmdBHHgkRVlVIRhHME5HrRKRxsF0L/BLkFES/ZnGciBgxwpI6Q+kgOXCgrQRSDjvKHapUsSiiDz+E//wnbmmylCuusFIdU6fCQQfBN99EMk0yiuAcYAfghWBrFOyrDJwZiVSOkwaGDoVWraBZsxQHWrwYXnrJHn//jJ10wNwlZ54JN9+cV4FU4XLaaeY3WLrUQrMiYJPBcqq6CLhMRDa3j7oi4fDMSKRynIiZMcMcmin16S1i+HALGXWz0EaIwKOPWmWHzp3NmVyjxqavc4px4IH2w9t++0iG3+SKQET2FpEpWHbwlyIySUTCsKo6TmzssINZczp1CmGwQYOgZUurLulsRN269rM+6qi4JclyGjUKIdGlZJIZ9XHgSlUdByAibYEn2LiAnONkDbVrh/QA/8UX9qT2wAMhDJa7HHOMbU5mkoyPoFaREgBQ1fFArcgkioq1ay0ky2va5T2ffWY5X8uXhzDYoEHmFwhlaZH7fPKJBcAsWRK3JE4iySiC70TkRhFpEmx9gO+iFix0Bg2yOLbTT4dFi+KWxomRfv3gH/8I4Zlg7Vp4+mnLHahXLxTZ8oH33rMQeSdzSEYRXADUx/oJjAbqAedHKVQkXHgh3H03jBkDe+8Nr70Wt0RODKxdC888Y4mbdeqkONjrr1vrQ3cSJ02bNtaLZciQzGpglu9sUhGo6m+q2ktVW6pqK1W9AuiTBtnCpVIluPpqW5tuvbXViv/73+GPP+KWzEkjb71lC8JQcgcGDbKSAMceG8Jg+cM//wmtW8PFF1v/Hid+KtqlIXvzB1q0sIJOvXtbyeCWLc3Z5+QFQ4daFEvK9+5Fi6zLlOcOlJuqVc2itnKl+9gzhYoqAglVinRTo4YFkI8dCytWWIzuHXdYCUonZ1GFZcusVH3K7WqHDTM7U0HBps91NqJZM3j/fbj99rglcaDsfgQllZ8GUwKfqeoOkUlVBqGXof71V+jRwwzHhxxixsumTcMb38k4CgtD6FjYsqUN4iXRU2b+fFi1Cpo0iVuS3KaiZagnAROD18RtIrAmbCFjo25dywx9+mmLCd9nH8t+8TDTnGPZMntNWQl89pnVSzg/+2ImMo316+GIIyz61gvTxUep/xKq2lRVdwpei287pVPIyBEx7+Hnn1vxmQsusFBTDzPNGRYsML/ugAEhDDZ4sNmWzj47hMHym8qV4dZbrXT1XXfFLU3+UqoiEJEmZV0oRizmocho3BjefttKJb78soWZvvHGpq9zMp6RI8380KZNigMV5Q6cfHL6+vDmOGefbdutt3rcRlyUtUi+W0RGi0hXEdlTRLYRkUYicqSI9AXeB5qnSc70UbkyXHPNhjDTY4+1dnEeZprVDB1qVr+Uew+8+qrVhffcgVB5+GFbsXXpEnkPFqcEyjINnQHcCDQDHgYmAC8CFwFfA0eq6lvpEDIW9t3XHIFXXAEPPWQmI++/l5XMmmWmh9ByB7bbzgvnhExRYbq99oLVq+OWJv8o022mql+p6j9Vta2qNlPV/VT1HFV9WlVXpUvI2KhRwzoEvfWWeRoPOAD+9S8PM80yhg0zN1DK5YAWLoRXXrHH1oiqQOYzRx1lJrwtt4xbkvwj1fiJ/KB9e4soOu00a8bati18/33cUjlJUlBgD/I77pjiQMOGWWiL5w5Eyjff2L+aF6ZLH64IkqVuXettOGSIRRe1aGHRIx5mmvE0agRdu4Yw0MCBsP/+sOeeIQzmlMbSpdbwrWfPuCXJH1wRlAcRKynw+eew337mMDzjDGtV6GQkAwfCc8+FMNDUqZY/4E7iyNl/f7jpJnPwjxwZtzT5QTIdykREOovITcHnRiKSahBedtO4MbzzDvz73/bo4mGmGcn69Vbg7KmnQhhs0CDPHUgjN9xgob6XXAI//xy3NLlPMiuCR4CDgCJX23Isiii/qVwZrr0WPv4YttrKwkx79fLYtwxi3DirbplytNCaNfZ4esopZiJ0IqdKFbPCrloFffvGLU3uk4wiOEBV/w6sAitLDaRaslls36cAACAASURBVCt32G8/CzO9/HJ48EELM50yJW6pHOzevfnmcOKJKQ706quWZe5mobSy22620L733rglyX2SUQRrRaQyoAAiUh8ojFSqbKNmTet9+Oab5uk64ADLl/cw09hYudI6k3bsaL+elBg4EBo08O7rMXDYYfb7W7ECfvopbmlyl2QUwQPA88A2InIH8D/gzkilylaOOsrCTDt0sF6I7dtbkRsn7fzwA2y/fQhmoV9+sa52njsQG6r2r9Sxoxemi4pkOpQNBa4F/gXMA05R1Wc3dZ2INBORqQnbMhG5otg5bUVkacI5N1X0i2QMdetaqMPAgZbO2rq1ZyTHwO67w/Tp1ig9JYYNs5Wdm4ViQwSuusqqvtzpj6CRUGo/gj9PEGlU0n5V/THpScy0NBfzN/yQsL8tcLWqJm3FDb0fQZRMmgSnnmoZqU8+CeecE7dEecHKlXbzqFEjxYFULV9ks81MqTux0rmzpfJ8+KGFmDrlo6L9CIoYA7wSvL4NfAeUt/N7O2BWohLIC1q1Mkfy/vubjeKaa9xvkAaGDLFyQCnblKdMMVOfrwYygoceMldNly5eAzJskjEN7a2q+wSvuwJtgA/LOc/ZwPBSjh0kIp+JyGsiUmLKpoh0F5GJIjJx4cKF5Zw6ZrbZxlpi9ugB99wDxx9vXdGcyBg61G4YO6RaJH3QIKhe3XpbOrGz5Zb2K9lmG4vJcMJjk6ahEi8S+UJV907y3GrAz8CeqvpLsWN1gEJVXSEixwP3B8qmVLLKNFSc/v1NITRqBC+8EEJNZKc4P/5o+X59+0KfPikMtHq1eZuPOsrsEU7GoGqmP6d8pGQaEpErE7arRWQYdmNPluOAycWVAICqLlPVFcH7V4GqIlKvHGNnFxddBOPHw++/w4EHwvPPxy1RzjFsmL2m7I4ZM8ZWbt6OMuMQsWC8iy7yxXVYJOMj2Dxhq475CjqUY45OlGIWEpHtREy3B2UrKgG5Xbjn4IPNb7DnnlZi8eabrZu6EwpDh8JBB8FOqTZTHTTIVgTt24chlhMyc+ZYzce//z1uSXKDTQZGq+qtFR1cRGoBRwEXJ+y7JBj3MaAjcKmIrANWAmdrRWxV2UbDhvDuu3DppXDbbVbQbMgQqFMnbsmyGlULzkq5scn8+ZZNfM01VkrEyThatrRnqBtvtK6hKfeayHOSCR99mSCrOIGlwETg8XQ3qMlqH0FxVK0sxZVXWj79Cy/YqxMv//d/cPXVloiw++5xS+OUwrp1lnk8Y4YFd6UcHJDjpBo++h2wAugXbMuwwnO7BZ+diiJihereesuMnm3awGvljcx1wKxr11xjFcJTQtXMQgce6Eogw6lSxSrLrllj9R+dipOMIjg4aE/5crB1BvYPCtG1jFi+/OBvfzO/QZMmcMIJVqcoDyxkYTJhgkXnTpuW4kCffmqDuJM4K9h1V4u5uP/+uCXJbpJRBLUTs4uD97WDj2sikSofadIE3n8fzjzT6hR16mTRRU5SDBsGtWpZmaeU6NfPBnKjc9Zw9NFQv76Zin7ZKDbRSYZkqmhdBfxPRGYBAjQFegSO4MFRCpd31KoFw4dbaet//MOMny+8YErCKZU1a+DZZ61dQK1aKQy0fLn9/M86y+pXO1nFqaean/+DD6Bq1bilyS6SySx+FdgVuAK4HGimqmNU9XdV/W/UAuYdInDddRbHPnu2Fa0bNy5uqTKa116D334LodLoM8/YKqxbt1DkctJLly5mYX366bglCZ/CwhDMnmWQbM/iVsCeQAvgTBEJoxW4UxbHHWflFrfZxrJbH3jA/QalsHgxNG8eQruA/v0tv+OAA0KRy0kvZ5xhi+k778y9kl6VKlny3PTpEY2/qRNEZAhwD3AosH+wlRiC5ITMbrtZ1csTTrAOaBdcYL37nL9wwQXw5Zcptgv44gtrO9qtm9cvyFJErKzIzJm52fT+8MPtgScKklkRtAYOUdUeqnpZsPWKRhxnI+rUsbCIm26ysMYjjoC5c+OWKmP49deQas/072/N6Tt3DkUuJx5OOcUWdUWlRnKBMWMsynz58ujmSEYRTAO2i04EZ5NUqgS33mq9F7/80vwGH3wQt1QZwVlnhWASWrXKMrtPPx223joUuZx4qFQJXnkld8p4qVoG9ZgxIbRcLYNkFEE94CsReUNEXiraohPJKZXTTjNT0WabQdu29hSbx8ybB++8Y7WFUuK558zbfNFFocjlxEuTJmYmXLky+91qr71m/a1uuCHaTqnJDH1LdNM75WavvSzpqVMns2dPmQL33WdmjTxjxAiLpkg5WqhfP6tS17ZtGGI5GcC0adCuHQwYYC62bETVSpE1bmwRUVGSTPjou8BsoGrw/lPAm/DGSd26tla8+mp45BGrkLlgQdxSpZ2hQ634WEqVIL791kqDX3SR2RWcnKBZM2tV2rdv9q4Kxo61+IV//CP657xkooa6AaOAx4NdDYEXohTKSYIqVeDuuy1o+tNPzW8QVWxZBvL117ZkTtm3++STVmHU21HmFFWrwvXX2430nXfilqZiNG0KPXum508zmeqjU7H2lB+r6n7BvqQ7lIVNTlUfDYvJk20dfNRRuRk3VwKrV8Prr1ttuG23reAga9fCjjvaIC/4s02usWqVWfyaNfOcTEi9+uhqVf2zppCIVGHjstROnLRsCV27wosvmtMzD6he3eoKVVgJgIWX/PKLO4lzlBo1rCrp+PH2rJRN3HwzfPZZ+uZLRhG8KyI3ADVF5CjgWeDlaMVyyk1BgRXdeeaZuCWJnClT4JZbQmhT2L+/NQk69tgwxHIykO7dzTS0335xS5I8EyaYkzidq5hkFMH1wELgC6zT2KtAKm3BnSjYbz/LpBmc+3UABwyAf/87xXC6n34y29L550cbl+fEymabWZV3kexxGvfta5VlundP35zJRA0VYs7hHqraUVX75UU7yWxDxFYFH31kntQcZe1aW/ScdFKKnT0HDrQ7w4UXhiabk7nccksIYcZp4KOPrE/V1VebEksXpSoCMW4RkUXA18DXIrJQRG5Kn3hOuejc2UIgn3oqbkkiY+xYWLgwxX/q9estWqh9ey/xnSeIWIXxlDvYRUzfvpbcfuml6Z23rBVBb+AQrBtZXVWtCxwAHCIivdMinVM+GjSwLh1DhlimVQ7y9NOw1VZWnLXCjB0LP/7o5abziF69rMXEnXfGLUnpFBZCixZw441Qu/amzw+TshRBF6CTqn5ftENVvwM6A16GOlMpKDD79/jxcUsSCZUr28InpQSbfv2gXj04+eTQ5HIym622spj8kSMz13JaqZIpqssvj2HuMo5VVdVFxXeq6kLA+/9kKh06mPE8R53GTz2VYn/aBQsszLagwGJQnbyhd28LKc3EVcFXX1mxgLi8r2UpgrL6EXuv4kylZk3rezx6NKxYEbc0oVLUjzalktODB1tzW3cS5x3161vE8PXXxy3Jxtx0E5xzDixbFs/8ZSmCFiKyrIRtORBLVrGTJAUF1nJx9Oi4JQmNBQtghx3goYdSGETV7gSHHhpdhw8noznnnMz71U+bZv+qvXrBFlvEI0OpikBVK6tqnRK2zVXVTUOZzCGHwM4751T00LPP2oP8EUekMMiECfDNN55JnOd8/z2cfba50jKBO+4w5/AVV8Qng5dbzEVErOTEuHEWHZMDDB0Ke+9tW4Xp39/8J2ecEZpcTvYhYk/gd98dtyQwY4blxfz97/H2RHJFkKt06WKmkCFD4pYkZb77Dj78MMXcgd9+s2XFueemN1PHyTiaNLF/j379YP78eGWZM8dak195ZbxyuCLIVZo2tW7XgwdnT259Kfzvf/baqVMKgwwbZuUoPXfAwWr8r1kD994brxzt21v1+G22iVcOVwS5TEGBNV756KO4JakQRU9r55xjJQIaNargQKr2+NeyZXZVH3MiY9ddzU/wyCOweHE8MowbZ+XUU4qCCwlXBLlMx44WTpplTuMZM6y94L77WjhdlSpWlrfCTJpkNX19NeAkcMMNlmQWR2O677+39iF9+6Z/7pLIK0WwcCEsXx63FGmkTh1reD9ihJlFMpxff7XIib33NnPQ1VeHlPPVr5/5BVKyLTm5xp57wl13WdZxurnrLsuST3dNodKITBGISDMRmZqwLRORK4qdIyLygIjMFJHPRaRlVPKsXWvlaE87zZZjeUNBASxZAi9ndguJOXNsuf7gg5br9e23ISmCFSvMP3DmmfEFaTsZzWuvWQ2rdPHjj1b89qKLrB1GJhCZIlDVr1V1X1XdF2gF/AE8X+y044Bdg6078GhU8lStat2Kxo61WjXr10c1U4Zx5JH215ahJSdmzrTXhg2hRw+YOhUeeyxE59nIkaYMPHfAKYWHH7aVaLoS8f/9b3u97rr0zJcM6TINtQNmqeoPxfZ3AJ5S4yNgSxFpEJUQXbtalMCoUXbTyfJgmuQoqtL2+usbajRkAF99ZRVE99oLZs82h1nfvinmCZRE//6WSnrwwSEP7OQKffqYw/jxx6Ofq7DQHnbOOy+F4IcISJciOBsYXsL+hkBift+cYN9fEJHuIjJRRCYuXLgwJUF69zYn0RNPwH33pTRU9lBQYEugYcPiloRFi8xBt88+lhvwr3/B9ttHNNmXX9okF12UGaEZTkZy4IEWxnnPPbByZbRzVapk/q///jfaecpL5IpARKoBJ2O9jiuEqj6hqq1VtXX9+vVTlun2281Z07lzykNlB82bw/77x24eWrYMdt/dTD8XX2xmod69UywpXRb9+5tNsKtXTXfKpk8fC1ceMCC6OX791TaRzMtpTMeK4DhgsqqWZJeYC+yY8HmHYF+kiJh9bpttzIk8YULUM2YABQUWQvnZZ2mdVhUmT7b3depYU+7PPjO7bL16EU68apWFzZ56asQTObnA4YfDWWel2P50E9x5p5UAW7o0ujkqSjoUQSdKNgsBvAR0DaKHDgSWquq8NMj0J3fcYdFEY8akc9YYOPtsezpO46pg2jQ45hho1Qo+/tj29ehhYXuR88IL9vjlTmInCUQsyrpLl2jGX7gQHn0UTjwxM4PXIlUEIlILOAp4LmHfJSJySfDxVeA7YCbQD+gRpTwlcdVVlmzasWOOrwy23tr+CocOtWVQhCxcaDf8Fi1g4kRrJNMyssDgUujXz4rKtGuX5omdbGbNGivPtW5duOPee6/5H/75z3DHDQ1VzaqtVatWGjYLFqg2a6a6xRaqU6eGPnzm8MILqqD6yiuRTbFmjeqOO6pWrqx62WWqixZFNlXpzJxp37Nv3xgmd7KZV16xP50hQ8Ibc/Fi1dq1Vc86K7wxKwIwUUu5r+ZVZnFp1K8Pb75pza3POCP8p4GM4bjjzF4esnlIFd55x16rVrVorC++gAceiKm07oABFp5x/vkxTO5kM8cdZxFtd9wRXq7RmDHWJ6pPn3DGiwJXBAGNGpkyGD7catvkJNWqWQW3F1+0sswh8PnnFnrXrp0NC3D66TF2gVq3ztI2jz8+c9I2nayhUiUz38yYAc89t+nzk6FLF5g1y3JmMhVXBAk0b26OTbCH5pDulZlF165mCH3mmZSGWbDAQkD3288SZB56yArFxc6rr8K8eV5gzqkwp58OzZpZmHmqSadF2cpNm6YuV5S4IiiBWbPsPnLSSfDHH3FLEzItW1rYTgrmofXrrWXkgAHWZ3XmTOuwVDUTGpj26wcNGtiKwHEqQOXKlnRaqVJqyfjLlsFOO5mJNNNxRVACO+9swTUffGC1yiIOskkvIpZT8NFH1r+3HKxcaZaXypUtsmLaNPMHxFG9sUTmzrUVwfnn57B9z0kHnTtb/st221V8jEcesQi6bKhu4oqgFM44w+J+x4yBCy6wGiE5Q+fO9rhTjj4F33xjqfg33WSfW7e25XNGMXCg/aIuuCBuSZwsp1Ile2b67bdyPy8BZhL6v/8z53Pr1uHLFzauCMrg4ovNTvj00xvaJeYEDRrA0UfbY30SGm7ECPOdzJ0Lhx2WBvkqQmEhPPmkea133jluaZwcQBUOPbRiOYmPPWZ1tW68MXy5osAVwSa44QZLijr88LglCZmCAiuMPn58qaesWmWJYZ06WUjdlCn2hJORvP22lTH1TGInJETgkkss0fS995K/bv1666tx1FFw0EHRyRcmrgg2gciGSKKxY62OWU7QoYMVVinDafzNN+YQvvpq0xc77ljqqfHTvz/UrWu1hRwnJC66CLbd1iwDyVK5Mrz/vimDbMEVQTl49FHo3h2erXAd1QyiZk3zhI8evVFHjqK6dPvsY8rg7rszJCKoNBYuhOeft9DYUHpbOo5Rs6Y9CL311oZ6WWVRFG66ww4Z6EMrA1cE5WDIEIsAOPdc+8PIegoKLOUxyJxZs8Y6Ne27rwXfQGY1zyiVIUMstMvNQk4EXHKJZci/+eamz330UWsKuGxZ9HKFiWiWtelq3bq1Tpw4Mbb5lyyxGPpZs6ysQps2sYmSOqrWKLhxY34Y8DZnngmffGK5AXffHWGfgDBRhT32sBjWDz6IWxonR1m0aNPVzFevhl12sVqH772Xeb2QRGSSqpYYw+QrgnKy5ZbW9XHbba0cRVYjAl27ouPGcWKLH5kxw9p43n9/ligBsJv/jBmeSexESpESKKtB4uDBMGeORQplmhLYFK4IKkCDBtYB8d5745YkBLp0QVS5ZLMhTJpk6fVZRb9+Vi3wzDPjlsTJcUaPtvJVX3218bG1a63xzAEHWLRQtuGKoIJss41p/VmzrJpBiq2U087cufDyy1gRlMMPp0etweyyc3aZCVm6FEaOtEJ6tWrFLY2T4xxxhK2U77xz42NDh8IPP2TnagBcEaTM/Pkwbpwpg+XL45YmOd54wxzCF1xgvmIKCpCZ3yYXFpFJDBtmdS/cSeykgXr14NJLzSQ8c+Zfj51+ui1Os7XElSuCFDnkEAsnnTIFTjnFHEaZyvr19sRy3HFm3vrf/4IH6Y4dLU4u5ub25aZ/f9NoRYkejhMxV11lodR33fXX/Ztvbs8j2bgaAFcEoXDiiVbm5p13zEoRVkOLMFm71myXt99uNdk++ighzrlOHTjtNKslsWpVrHImzeTJtmXzf5+TdWy3ncUljBhhlsn16y0387XX4pYsNVwRhESXLuY8njNno/ysjKBqVaubMmiQleTZbLNiJ3TtarGxL78ch3jlp39/qFHDkjocJ4306QPTp1sT+mefhZdeyv5y9Z5HEDJr1phDaf16SzWPk/XrzbF15JFmwtrkyY0bm6nllVfSIl+F+f132H57exQrRwVVxwmTwkJLYalSxTr1Vcrwx+qy8gi8aHvIVKtmTwcdOsBuu1kmctOmtm23XfqsGAsWWLXpt96yJewmFUHlynbBPfdYN45tt02LnBVi1ChL3fTcAScm1q/f0PJi+PDMVwKbIsvFz0yqV7cH1kcesXvrIYfY56Lm1cuXW0eve+6x2OTJk8NtizlhgrWQfO89i2S4++4kLywosL/wYcPCEyYK+vUzB8ehh8YtiZOnVK5sZer33dd6l2Q7bhqKkJUrrTLy99/b1qqVNXf5+mt7XbLkr+c//rgVtfvhB2tvt9NOG1YTTZpYYM+m+OADK5m9005mv2zRopxCt2lj9q2pU8t5YZqYPt3W43ffbdXAHCcmim6d2RKr4KahmKhZE5o3ty2RZs1sBbBkyQYl8f33G8w3s2bZaqJ4AM9rr8Gxx1p10Gef3aAkmja1aodVq5qCueMOi3euU6cCQhcUQM+eNkm5tUga6N/fvmjXrnFL4uQ52aIAksFXBBlKYaGZ6hMVRUGB9QR46ikLAU1sLla7tpXcadgwxYkXL7Ykg549M6+GxurVpvHats2RWuCOkz7KWhG4IshS1q6Fn37aoCTmzrW8sD33DGHw006zzhpz52ZWE/iRI+Gss6zq3zHHxC2N42QVbhrKQapWNT/ATjtFMHhBgTV6eeMNOOGECCaoIP37W4OE9u3jlsRxcgqPGnI25rjjrLBKJpWc+P57i4W98ML4EzQcJ8dwReBsTLVqVivjxRfDjWtNhQEDLFj7/PPjlsRxcg5XBE7JdO1qYaTPPBO3JLBunSmCY481b7njOKHiisApmZYtzfOcCSUcXn8dfv7Zy007TkREqghEZEsRGSUiM0RkuogcVOx4WxFZKiJTg+2mKOVxyoGIOY0//BC++SYeGQoLrUrehRdaSOuJJ8Yjh+PkOFGvCO4HXlfV3YEWwPQSzpmgqvsG220Ry+OUh86dzS4fx6pg4kTLsDv/fNh5Z8umq1o1/XI4Th4QmSIQkS2Aw4EnAVR1jaouKfsqJ6No0ACOPhqGDPlr9lqULFpkdTbatLFIocGDrYNOJmY5O06OEOWKoCmwEBgoIlNEpL+IlNRY9iAR+UxEXhORMNKhnDDp2hV+/BHGj492nnXr4OGHYdddzTF8xRVWlKlr1+wv7eg4GU6U/2FVgJbAo6q6H/A7cH2xcyYDjVW1BfAg8EJJA4lIdxGZKCITF2Zbl/hs55RTrGhRlOahCROsIl/Pnvb6+edW3mKLLaKb03GcP4lSEcwB5qhqUUf0UZhi+BNVXaaqK4L3rwJVRaRe8YFU9QlVba2qrevXrx+hyM5G1KwJZ55pPQDCbr02d651GDv8cMtXePZZSxrbY49w53Ecp0wiUwSqOh/4SUSKOuO2A75KPEdEthOxGn4i0iaQZ3FUMjkVpKDAuoI991w4461ZA//5j5VhHT0abrzRKuZ17JhbJR0dJ0uIutbQZcBQEakGfAecLyKXAKjqY0BH4FIRWQesBM7WbKuClw8ccohF7gwenHr559dfh8svt5DUk0+G++6LqGCS4zjJEqkiUNWpQPFqd48lHH8IeChKGZwQEDEFcMst5jhu1Kj8Y3z3HVx5pZWt2HVXePVVq2nkOE7seDiGkxxdulhLpqefLt91f/wBN91kdv+xY+Guu+CLL1wJOE4G4YrASY6mTc2pO3jwhh59ZaFq9v/mzaFvX+tx8PXXcN111tTZcZyMwRWBkzwFBWbb//jjss/76is46ihz/m6xheUgDBsWQvs0x3GiwBWBkzwdO1o4aWl9CpYtg6uusizgSZPgwQdh8mQ44oj0yuk4TrlwReAkT506ZuIZMQJWrdqwv7DQlMNuu1kU0Hnn2cqhZ8/ManXpOE6JuCJwykfXrrBkCbzyin2ePBkOPdRu/k2amNmoXz/wxD/HyRr8cc0pH+3ama3/0UctCuiJJ+ymP2CA+RC8LpDjZB3+X+uUj8qVrTz1O+9YM/nLL7dooPPPdyXgOFmKrwic8nPFFZYf0L077LVX3NI4jpMirgic8rPddvDAA3FL4ThOSPha3nEcJ89xReA4jpPnuCJwHMfJc1wROI7j5DmuCBzHcfIcVwSO4zh5jisCx3GcPMcVgeM4Tp4j2dYiWEQWAj9U8PJ6wKIQxck0cvn7+XfLXnL5+2XTd2usqiVWg8w6RZAKIjJRVYv3UM4Zcvn7+XfLXnL5++XKd3PTkOM4Tp7jisBxHCfPyTdF8ETcAkRMLn8//27ZSy5/v5z4bnnlI3Acx3E2Jt9WBI7jOE4xXBE4juPkOXmjCETkWBH5WkRmisj1ccsTFiKyo4iME5GvRORLEbk8bpnCRkQqi8gUEXklblnCRkS2FJFRIjJDRKaLyEFxyxQWItI7+JucJiLDRaRG3DKlgogMEJEFIjItYV9dEXlLRL4NXreKU8aKkheKQEQqAw8DxwF7AJ1EZI94pQqNdcBVqroHcCDw9xz6bkVcDkyPW4iIuB94XVV3B1qQI99TRBoCvYDWqroXUBk4O16pUmYQcGyxfdcDb6vqrsDbweesIy8UAdAGmKmq36nqGmAE0CFmmUJBVeep6uTg/XLsRtIwXqnCQ0R2AE4A+sctS9iIyBbA4cCTAKq6RlWXxCtVqFQBaopIFWAz4OeY5UkJVX0P+LXY7g7A4OD9YOCUtAoVEvmiCBoCPyV8nkMO3SyLEJEmwH7Ax/FKEir/Ba4FCuMWJAKaAguBgYHpq7+I1IpbqDBQ1bnAPcCPwDxgqaq+Ga9UkbCtqs4L3s8Hto1TmIqSL4og5xGR2sBo4ApVXRa3PGEgIicCC1R1UtyyREQVoCXwqKruB/xOlpoWihPYyjtgym57oJaIdI5XqmhRi8XPynj8fFEEc4EdEz7vEOzLCUSkKqYEhqrqc3HLEyKHACeLyGzMnHekiDwdr0ihMgeYo6pFK7hRmGLIBdoD36vqQlVdCzwHHByzTFHwi4g0AAheF8QsT4XIF0XwKbCriDQVkWqY0+qlmGUKBRERzMY8XVXvjVueMFHVf6jqDqraBPudvaOqOfNUqarzgZ9EpFmwqx3wVYwihcmPwIEislnwN9qOHHGEF+MloCB4XwC8GKMsFaZK3AKkA1VdJyI9gTew6IUBqvplzGKFxSFAF+ALEZka7LtBVV+NUSYneS4DhgYPKN8B58csTyio6sciMgqYjEW2TSHLyzGIyHCgLVBPROYANwN3ASNF5EKsPP6Z8UlYcbzEhOM4Tp6TL6Yhx3EcpxRcETiO4+Q5rggcx3HyHFcEjuM4eY4rAsdxnDzHFYGTE4jI1iIyNdjmi8jchM8fRDTnfiLyZPD+PBFREWmfcPyUYF/HCo7fVkQOTvg8qKSxRKS+iLxekTkcB/Ikj8DJfVR1MbAvgIjcAqxQ1XsinvYG4PaEz19giW9jg8+dgM9SGL8tsAIoU5Gp6kIRmScih6jq+ynM5+QpviJwch4RWRG8thWRd0XkRRH5TkTuEpFzReQTEflCRHYOzqsvIqNF5NNgO6SEMTcH9lHVxBv9BKCNiFQNaj/tAkxNuKZdUFzui6C2ffVg/2wRuVVEJgfHdg8KCF4C9A5WNYcFwxwuIh8E8ieuDl4Azg3rZ+bkF64InHyjBXaDbY5lZO+mqm2wMteXBefcD9ynqvsDp1NyCezWwLRi+xRbn/uADgAAAaNJREFUDRyDFVz7s4xJ0JRlEHCWqu6NrcYvTbh2kaq2BB4FrlbV2cBjgRz7quqE4LwGwKHAiVhWaxETgcNwnArgisDJNz4NejisBmYBRaWRvwCaBO/bAw8FJTteAuoET/iJNMBKSBdnBGYeOhsYnrC/GVaE7Zvg82CsF0ERRcUCJyXIURIvqGqhqn7FX0seL8CqfDpOuXEfgZNvrE54X5jwuZAN/w+VgANVdVUZ46wENmq9qKqfiMjewB+q+o3VWyuXXOsp+/8yUf7EwWsEMjlOufEVgeNszJtsMBMhIvuWcM50zAdQEtdjjuREvgaaiEjRNV2Adzchx3Jg801Ka+zGxqYqx0kKVwSOszG9gNYi8rmIfIX5FP6Cqs4AtgicxsWPvaaq44rtW4VVFn1WRL7AViCPbUKOl4FTizmLS+NvwJhNnOM4JeLVRx2ngohIb2C5qsbeT1lE3gM6qOpvccviZB++InCcivMof7XZx4KI1AfudSXgVBRfETiO4+Q5viJwHMfJc1wROI7j5DmuCBzHcfIcVwSO4zh5jisCx3GcPOf/AbsBS2qGophUAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig = plt.figure()\n",
    "ax = plt.axes()\n",
    "ax.set_title('Region: 3 \\n MSLE: '+str(R3_msle))\n",
    "ax.set_ylabel('Dengue (Log)')\n",
    "ax.set_xlabel('Time (Month)')\n",
    "plt.plot(predict_R3_dengue, 'b--')\n",
    "plt.plot(R3_test_dengue, 'r-')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6894,
     "status": "ok",
     "timestamp": 1600944492214,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "XK2BmOAtwOcK"
   },
   "outputs": [],
   "source": [
    "predict = pd.DataFrame(predict_R3_dengue)\n",
    "actual = pd.DataFrame(R3_test_dengue)\n",
    "R3_eval = pd.concat([predict, actual], axis=1)\n",
    "R3_eval.columns = ['Predict', 'Actual']\n",
    "R3_eval.index = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']\n",
    "#R3_eval.to_csv('R3_eval.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6887,
     "status": "ok",
     "timestamp": 1600944492215,
     "user": {
      "displayName": "Bryan Zafra",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14Gg_PwZ4BbMV-4NtGpXP_NDm8G2Id3zhdwFztewNfQ=s64",
      "userId": "14572305043179199752"
     },
     "user_tz": -480
    },
    "id": "rhygB9SjPMqJ"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Region3.ipynb",
   "provenance": [
    {
     "file_id": "1eFe2tHTrIO9MbXL__mtqYiKmBCFxGcSW",
     "timestamp": 1600943936107
    },
    {
     "file_id": "1ryQzRHVo_3AhAYbmxPXKA2UmQa-ka387",
     "timestamp": 1600943299051
    },
    {
     "file_id": "1LR3-ITBJFJehhWTNr1Xrd6AoeqkK2fdl",
     "timestamp": 1600942956635
    },
    {
     "file_id": "1nLXQRJKkt4KSfm8XUlqfdezhdvbmlKkJ",
     "timestamp": 1600921226564
    },
    {
     "file_id": "1rczDnJQLfdoVdf_4-AvBEESkZ5cR4opI",
     "timestamp": 1600920727380
    },
    {
     "file_id": "1v0TcWQxp8YtU3EJItyQneweX7vu3wldv",
     "timestamp": 1600919671140
    },
    {
     "file_id": "1o25i2xAR0szppQdGxxqC0EFmXBCbLggT",
     "timestamp": 1600598548116
    },
    {
     "file_id": "1XSseqmpktSfnyvxevYWPUMhQwLqOWlHw",
     "timestamp": 1600343682319
    }
   ]
  },
  "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}