TensorFlow_TimeSeriesGenerator Example.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Playground.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "ujsyT9u9udnJ"
      },
      "source": [
        "import os\r\n",
        "os.environ['KAGGLE_USERNAME'] = \"\"\r\n",
        "os.environ['KAGGLE_KEY'] = \"\""
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "11sHAoYlu_wq",
        "outputId": "0eb0f951-fd74-4fe7-82d3-870fcf09a280"
      },
      "source": [
        "!kaggle datasets download -d robikscube/hourly-energy-consumption"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading hourly-energy-consumption.zip to /content\n",
            " 61% 7.00M/11.4M [00:00<00:00, 30.0MB/s]\n",
            "100% 11.4M/11.4M [00:00<00:00, 38.0MB/s]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YkkjRh4fwCN7",
        "outputId": "3ef41f33-6065-4f7e-d205-bdfe1bfceb87"
      },
      "source": [
        "!unzip /content/hourly-energy-consumption.zip"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Archive:  /content/hourly-energy-consumption.zip\n",
            "  inflating: AEP_hourly.csv          \n",
            "  inflating: COMED_hourly.csv        \n",
            "  inflating: DAYTON_hourly.csv       \n",
            "  inflating: DEOK_hourly.csv         \n",
            "  inflating: DOM_hourly.csv          \n",
            "  inflating: DUQ_hourly.csv          \n",
            "  inflating: EKPC_hourly.csv         \n",
            "  inflating: FE_hourly.csv           \n",
            "  inflating: NI_hourly.csv           \n",
            "  inflating: PJME_hourly.csv         \n",
            "  inflating: PJMW_hourly.csv         \n",
            "  inflating: PJM_Load_hourly.csv     \n",
            "  inflating: est_hourly.paruqet      \n",
            "  inflating: pjm_hourly_est.csv      \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DZd7U7J0wPxL"
      },
      "source": [
        "import pandas as pd\r\n",
        "import numpy as np\r\n",
        "import tensorflow as tf\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "\r\n",
        "from sklearn.model_selection import train_test_split\r\n",
        "from keras.preprocessing.sequence import TimeseriesGenerator\r\n",
        "from sklearn.preprocessing import MinMaxScaler"
      ],
      "execution_count": 72,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "g8KqJIg1wWUp",
        "outputId": "ea1f7cf6-2c76-48f9-9b37-45c521341b28"
      },
      "source": [
        "df = pd.read_csv('/content/AEP_hourly.csv', parse_dates=[0])\r\n",
        "df"
      ],
      "execution_count": 468,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Datetime</th>\n",
              "      <th>AEP_MW</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2004-12-31 01:00:00</td>\n",
              "      <td>13478.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2004-12-31 02:00:00</td>\n",
              "      <td>12865.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2004-12-31 03:00:00</td>\n",
              "      <td>12577.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2004-12-31 04:00:00</td>\n",
              "      <td>12517.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2004-12-31 05:00:00</td>\n",
              "      <td>12670.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>121268</th>\n",
              "      <td>2018-01-01 20:00:00</td>\n",
              "      <td>21089.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>121269</th>\n",
              "      <td>2018-01-01 21:00:00</td>\n",
              "      <td>20999.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>121270</th>\n",
              "      <td>2018-01-01 22:00:00</td>\n",
              "      <td>20820.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>121271</th>\n",
              "      <td>2018-01-01 23:00:00</td>\n",
              "      <td>20415.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>121272</th>\n",
              "      <td>2018-01-02 00:00:00</td>\n",
              "      <td>19993.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>121273 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                  Datetime   AEP_MW\n",
              "0      2004-12-31 01:00:00  13478.0\n",
              "1      2004-12-31 02:00:00  12865.0\n",
              "2      2004-12-31 03:00:00  12577.0\n",
              "3      2004-12-31 04:00:00  12517.0\n",
              "4      2004-12-31 05:00:00  12670.0\n",
              "...                    ...      ...\n",
              "121268 2018-01-01 20:00:00  21089.0\n",
              "121269 2018-01-01 21:00:00  20999.0\n",
              "121270 2018-01-01 22:00:00  20820.0\n",
              "121271 2018-01-01 23:00:00  20415.0\n",
              "121272 2018-01-02 00:00:00  19993.0\n",
              "\n",
              "[121273 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 468
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q_55Qimwzgx3"
      },
      "source": [
        "df = df[:-120273] "
      ],
      "execution_count": 469,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "id": "AbL-Fsqiz2_S",
        "outputId": "758394f0-be16-465b-a462-ce5ac3c307ee"
      },
      "source": [
        "df.plot(x='Datetime')\r\n",
        "plt.show()"
      ],
      "execution_count": 470,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H517pw8M41zD",
        "outputId": "f8c8e645-efc8-4dfd-eeec-8a1a9355ab0d"
      },
      "source": [
        "minmaxscaler = MinMaxScaler(feature_range = (0,1))\r\n",
        "scaled = minmaxscaler.fit_transform(df[['AEP_MW']])\r\n",
        "df[['AEP_MW']] = scaled"
      ],
      "execution_count": 471,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n",
            "/usr/local/lib/python3.6/dist-packages/pandas/core/indexing.py:1734: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  isetter(loc, value[:, i].tolist())\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "26p4kLYZ1Do5"
      },
      "source": [
        "df_train, df_test = train_test_split(df, test_size=0.1, shuffle=False)"
      ],
      "execution_count": 472,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eOMuXbNg3O4Y"
      },
      "source": [
        "df_train, df_val = train_test_split(df_train, test_size=0.2, shuffle=False)"
      ],
      "execution_count": 473,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dgNwBVI4PiXY"
      },
      "source": [
        "energy_train = df_train['AEP_MW'].values.reshape((len(energy_train)), 1)\r\n",
        "energy_val = df_val['AEP_MW'].values.reshape((len(energy_val)), 1)\r\n",
        "energy_test = df_test['AEP_MW'].values.reshape((len(energy_test)), 1)"
      ],
      "execution_count": 486,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BrLfOMHC37ty"
      },
      "source": [
        "train_sets = TimeseriesGenerator(energy_train, energy_train, length=2, batch_size=1)\r\n",
        "val_sets = TimeseriesGenerator(energy_val, energy_val, length=2, batch_size=1)\r\n",
        "test_sets = TimeseriesGenerator(energy_test, energy_test, length=2, batch_size=1)"
      ],
      "execution_count": 487,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dvsxoRn34Tpa",
        "outputId": "be78a7ce-4c1b-4934-84b2-019f4c2f9a93"
      },
      "source": [
        "for i in range(1):\r\n",
        "\tx, y = test_sets[i]\r\n",
        "\tprint('%s => %s' % (x, y))"
      ],
      "execution_count": 489,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[[16202.]\n",
            "  [16062.]]] => [[15836.]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dop4BwBS5d_e",
        "outputId": "ee43b949-9190-443f-c6ee-e6a6cb7472b6"
      },
      "source": [
        "tf.keras.backend.clear_session()\r\n",
        "model = tf.keras.models.Sequential([\r\n",
        "  tf.keras.layers.LSTM(128, input_shape=(2,1)),\r\n",
        "  tf.keras.layers.Dense(64, activation=\"relu\"),\r\n",
        "  tf.keras.layers.Dense(32, activation=\"relu\"),\r\n",
        "  tf.keras.layers.Dense(1),\r\n",
        "])\r\n",
        "optimizer = tf.keras.optimizers.SGD(lr=1.0000e-04, momentum=0.9)\r\n",
        "model.compile(loss=tf.keras.losses.Huber(),\r\n",
        "              optimizer=optimizer,\r\n",
        "              metrics=[\"mae\"])\r\n",
        "history = model.fit_generator(train_sets, epochs=150, validation_data=val_sets)"
      ],
      "execution_count": 434,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  warnings.warn('`Model.fit_generator` is deprecated and '\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/150\n",
            "718/718 [==============================] - 4s 3ms/step - loss: 0.0894 - mae: 0.3830 - val_loss: 0.0111 - val_mae: 0.1249\n",
            "Epoch 2/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0198 - mae: 0.1589 - val_loss: 0.0167 - val_mae: 0.1537\n",
            "Epoch 3/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0124 - mae: 0.1244 - val_loss: 0.0208 - val_mae: 0.1697\n",
            "Epoch 4/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0142 - mae: 0.1345 - val_loss: 0.0215 - val_mae: 0.1722\n",
            "Epoch 5/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0131 - mae: 0.1306 - val_loss: 0.0215 - val_mae: 0.1723\n",
            "Epoch 6/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0130 - mae: 0.1281 - val_loss: 0.0213 - val_mae: 0.1717\n",
            "Epoch 7/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0129 - mae: 0.1279 - val_loss: 0.0212 - val_mae: 0.1710\n",
            "Epoch 8/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0140 - mae: 0.1321 - val_loss: 0.0212 - val_mae: 0.1710\n",
            "Epoch 9/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0125 - mae: 0.1245 - val_loss: 0.0210 - val_mae: 0.1705\n",
            "Epoch 10/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0131 - mae: 0.1285 - val_loss: 0.0214 - val_mae: 0.1720\n",
            "Epoch 11/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0120 - mae: 0.1208 - val_loss: 0.0212 - val_mae: 0.1712\n",
            "Epoch 12/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0131 - mae: 0.1314 - val_loss: 0.0211 - val_mae: 0.1708\n",
            "Epoch 13/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0126 - mae: 0.1261 - val_loss: 0.0210 - val_mae: 0.1705\n",
            "Epoch 14/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0123 - mae: 0.1247 - val_loss: 0.0216 - val_mae: 0.1728\n",
            "Epoch 15/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0138 - mae: 0.1323 - val_loss: 0.0211 - val_mae: 0.1705\n",
            "Epoch 16/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0117 - mae: 0.1213 - val_loss: 0.0206 - val_mae: 0.1688\n",
            "Epoch 17/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0122 - mae: 0.1237 - val_loss: 0.0203 - val_mae: 0.1676\n",
            "Epoch 18/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0120 - mae: 0.1250 - val_loss: 0.0206 - val_mae: 0.1686\n",
            "Epoch 19/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0125 - mae: 0.1249 - val_loss: 0.0197 - val_mae: 0.1648\n",
            "Epoch 20/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0121 - mae: 0.1233 - val_loss: 0.0199 - val_mae: 0.1656\n",
            "Epoch 21/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0117 - mae: 0.1226 - val_loss: 0.0196 - val_mae: 0.1645\n",
            "Epoch 22/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0122 - mae: 0.1231 - val_loss: 0.0198 - val_mae: 0.1653\n",
            "Epoch 23/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0115 - mae: 0.1216 - val_loss: 0.0198 - val_mae: 0.1653\n",
            "Epoch 24/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0120 - mae: 0.1211 - val_loss: 0.0195 - val_mae: 0.1639\n",
            "Epoch 25/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0123 - mae: 0.1244 - val_loss: 0.0196 - val_mae: 0.1644\n",
            "Epoch 26/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0116 - mae: 0.1224 - val_loss: 0.0193 - val_mae: 0.1633\n",
            "Epoch 27/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0126 - mae: 0.1277 - val_loss: 0.0190 - val_mae: 0.1617\n",
            "Epoch 28/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0119 - mae: 0.1222 - val_loss: 0.0193 - val_mae: 0.1632\n",
            "Epoch 29/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0120 - mae: 0.1236 - val_loss: 0.0193 - val_mae: 0.1631\n",
            "Epoch 30/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0112 - mae: 0.1222 - val_loss: 0.0193 - val_mae: 0.1631\n",
            "Epoch 31/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0107 - mae: 0.1162 - val_loss: 0.0185 - val_mae: 0.1598\n",
            "Epoch 32/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0111 - mae: 0.1197 - val_loss: 0.0188 - val_mae: 0.1610\n",
            "Epoch 33/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0122 - mae: 0.1258 - val_loss: 0.0182 - val_mae: 0.1584\n",
            "Epoch 34/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0102 - mae: 0.1132 - val_loss: 0.0186 - val_mae: 0.1599\n",
            "Epoch 35/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0107 - mae: 0.1155 - val_loss: 0.0182 - val_mae: 0.1584\n",
            "Epoch 36/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0105 - mae: 0.1148 - val_loss: 0.0180 - val_mae: 0.1576\n",
            "Epoch 37/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0108 - mae: 0.1176 - val_loss: 0.0180 - val_mae: 0.1574\n",
            "Epoch 38/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0109 - mae: 0.1163 - val_loss: 0.0174 - val_mae: 0.1547\n",
            "Epoch 39/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0113 - mae: 0.1208 - val_loss: 0.0170 - val_mae: 0.1531\n",
            "Epoch 40/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0108 - mae: 0.1172 - val_loss: 0.0169 - val_mae: 0.1524\n",
            "Epoch 41/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0103 - mae: 0.1131 - val_loss: 0.0171 - val_mae: 0.1533\n",
            "Epoch 42/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0114 - mae: 0.1194 - val_loss: 0.0173 - val_mae: 0.1540\n",
            "Epoch 43/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0102 - mae: 0.1133 - val_loss: 0.0169 - val_mae: 0.1522\n",
            "Epoch 44/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0111 - mae: 0.1174 - val_loss: 0.0166 - val_mae: 0.1512\n",
            "Epoch 45/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0103 - mae: 0.1168 - val_loss: 0.0162 - val_mae: 0.1492\n",
            "Epoch 46/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0095 - mae: 0.1105 - val_loss: 0.0161 - val_mae: 0.1485\n",
            "Epoch 47/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0099 - mae: 0.1126 - val_loss: 0.0156 - val_mae: 0.1466\n",
            "Epoch 48/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0093 - mae: 0.1074 - val_loss: 0.0158 - val_mae: 0.1471\n",
            "Epoch 49/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0099 - mae: 0.1129 - val_loss: 0.0155 - val_mae: 0.1459\n",
            "Epoch 50/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0100 - mae: 0.1130 - val_loss: 0.0152 - val_mae: 0.1442\n",
            "Epoch 51/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0094 - mae: 0.1086 - val_loss: 0.0149 - val_mae: 0.1427\n",
            "Epoch 52/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0100 - mae: 0.1133 - val_loss: 0.0148 - val_mae: 0.1422\n",
            "Epoch 53/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0098 - mae: 0.1121 - val_loss: 0.0147 - val_mae: 0.1419\n",
            "Epoch 54/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0095 - mae: 0.1102 - val_loss: 0.0138 - val_mae: 0.1373\n",
            "Epoch 55/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0086 - mae: 0.1025 - val_loss: 0.0147 - val_mae: 0.1419\n",
            "Epoch 56/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0091 - mae: 0.1079 - val_loss: 0.0136 - val_mae: 0.1362\n",
            "Epoch 57/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0083 - mae: 0.1031 - val_loss: 0.0133 - val_mae: 0.1349\n",
            "Epoch 58/150\n",
            "718/718 [==============================] - 2s 2ms/step - loss: 0.0090 - mae: 0.1074 - val_loss: 0.0136 - val_mae: 0.1360\n",
            "Epoch 59/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0077 - mae: 0.0974 - val_loss: 0.0139 - val_mae: 0.1376\n",
            "Epoch 60/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0084 - mae: 0.1022 - val_loss: 0.0128 - val_mae: 0.1317\n",
            "Epoch 61/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0083 - mae: 0.1006 - val_loss: 0.0124 - val_mae: 0.1299\n",
            "Epoch 62/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0086 - mae: 0.1051 - val_loss: 0.0128 - val_mae: 0.1319\n",
            "Epoch 63/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0084 - mae: 0.1018 - val_loss: 0.0122 - val_mae: 0.1286\n",
            "Epoch 64/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0074 - mae: 0.0969 - val_loss: 0.0118 - val_mae: 0.1266\n",
            "Epoch 65/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0075 - mae: 0.0974 - val_loss: 0.0117 - val_mae: 0.1261\n",
            "Epoch 66/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0077 - mae: 0.0989 - val_loss: 0.0115 - val_mae: 0.1246\n",
            "Epoch 67/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0070 - mae: 0.0942 - val_loss: 0.0110 - val_mae: 0.1221\n",
            "Epoch 68/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0068 - mae: 0.0920 - val_loss: 0.0115 - val_mae: 0.1246\n",
            "Epoch 69/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0074 - mae: 0.0946 - val_loss: 0.0105 - val_mae: 0.1192\n",
            "Epoch 70/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0076 - mae: 0.0972 - val_loss: 0.0105 - val_mae: 0.1191\n",
            "Epoch 71/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0070 - mae: 0.0955 - val_loss: 0.0105 - val_mae: 0.1190\n",
            "Epoch 72/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0064 - mae: 0.0897 - val_loss: 0.0100 - val_mae: 0.1160\n",
            "Epoch 73/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0067 - mae: 0.0911 - val_loss: 0.0096 - val_mae: 0.1135\n",
            "Epoch 74/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0066 - mae: 0.0924 - val_loss: 0.0100 - val_mae: 0.1161\n",
            "Epoch 75/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0060 - mae: 0.0867 - val_loss: 0.0096 - val_mae: 0.1137\n",
            "Epoch 76/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0061 - mae: 0.0866 - val_loss: 0.0090 - val_mae: 0.1097\n",
            "Epoch 77/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0061 - mae: 0.0877 - val_loss: 0.0090 - val_mae: 0.1100\n",
            "Epoch 78/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0058 - mae: 0.0840 - val_loss: 0.0080 - val_mae: 0.1035\n",
            "Epoch 79/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0055 - mae: 0.0830 - val_loss: 0.0084 - val_mae: 0.1060\n",
            "Epoch 80/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0053 - mae: 0.0804 - val_loss: 0.0084 - val_mae: 0.1060\n",
            "Epoch 81/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0053 - mae: 0.0820 - val_loss: 0.0081 - val_mae: 0.1037\n",
            "Epoch 82/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0056 - mae: 0.0822 - val_loss: 0.0081 - val_mae: 0.1037\n",
            "Epoch 83/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0055 - mae: 0.0832 - val_loss: 0.0075 - val_mae: 0.1000\n",
            "Epoch 84/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0050 - mae: 0.0799 - val_loss: 0.0070 - val_mae: 0.0958\n",
            "Epoch 85/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0054 - mae: 0.0818 - val_loss: 0.0070 - val_mae: 0.0964\n",
            "Epoch 86/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0051 - mae: 0.0799 - val_loss: 0.0069 - val_mae: 0.0957\n",
            "Epoch 87/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0050 - mae: 0.0777 - val_loss: 0.0070 - val_mae: 0.0960\n",
            "Epoch 88/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0042 - mae: 0.0720 - val_loss: 0.0067 - val_mae: 0.0941\n",
            "Epoch 89/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0047 - mae: 0.0758 - val_loss: 0.0065 - val_mae: 0.0923\n",
            "Epoch 90/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0046 - mae: 0.0748 - val_loss: 0.0061 - val_mae: 0.0891\n",
            "Epoch 91/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0044 - mae: 0.0736 - val_loss: 0.0066 - val_mae: 0.0932\n",
            "Epoch 92/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0045 - mae: 0.0736 - val_loss: 0.0056 - val_mae: 0.0850\n",
            "Epoch 93/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0043 - mae: 0.0730 - val_loss: 0.0056 - val_mae: 0.0853\n",
            "Epoch 94/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0043 - mae: 0.0710 - val_loss: 0.0055 - val_mae: 0.0843\n",
            "Epoch 95/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0045 - mae: 0.0715 - val_loss: 0.0056 - val_mae: 0.0849\n",
            "Epoch 96/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0037 - mae: 0.0665 - val_loss: 0.0051 - val_mae: 0.0805\n",
            "Epoch 97/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0039 - mae: 0.0682 - val_loss: 0.0054 - val_mae: 0.0832\n",
            "Epoch 98/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0035 - mae: 0.0644 - val_loss: 0.0051 - val_mae: 0.0808\n",
            "Epoch 99/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0038 - mae: 0.0674 - val_loss: 0.0047 - val_mae: 0.0769\n",
            "Epoch 100/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0038 - mae: 0.0688 - val_loss: 0.0046 - val_mae: 0.0758\n",
            "Epoch 101/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0034 - mae: 0.0654 - val_loss: 0.0049 - val_mae: 0.0790\n",
            "Epoch 102/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0038 - mae: 0.0668 - val_loss: 0.0048 - val_mae: 0.0778\n",
            "Epoch 103/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0034 - mae: 0.0637 - val_loss: 0.0044 - val_mae: 0.0740\n",
            "Epoch 104/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0037 - mae: 0.0673 - val_loss: 0.0042 - val_mae: 0.0718\n",
            "Epoch 105/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0037 - mae: 0.0657 - val_loss: 0.0041 - val_mae: 0.0711\n",
            "Epoch 106/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0031 - mae: 0.0606 - val_loss: 0.0039 - val_mae: 0.0694\n",
            "Epoch 107/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0033 - mae: 0.0626 - val_loss: 0.0039 - val_mae: 0.0687\n",
            "Epoch 108/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0032 - mae: 0.0610 - val_loss: 0.0036 - val_mae: 0.0666\n",
            "Epoch 109/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0035 - mae: 0.0625 - val_loss: 0.0039 - val_mae: 0.0689\n",
            "Epoch 110/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0578 - val_loss: 0.0037 - val_mae: 0.0675\n",
            "Epoch 111/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0031 - mae: 0.0606 - val_loss: 0.0036 - val_mae: 0.0658\n",
            "Epoch 112/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0568 - val_loss: 0.0035 - val_mae: 0.0656\n",
            "Epoch 113/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0033 - mae: 0.0619 - val_loss: 0.0033 - val_mae: 0.0635\n",
            "Epoch 114/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0034 - mae: 0.0636 - val_loss: 0.0035 - val_mae: 0.0650\n",
            "Epoch 115/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0027 - mae: 0.0550 - val_loss: 0.0035 - val_mae: 0.0650\n",
            "Epoch 116/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0030 - mae: 0.0591 - val_loss: 0.0032 - val_mae: 0.0617\n",
            "Epoch 117/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0033 - mae: 0.0613 - val_loss: 0.0032 - val_mae: 0.0622\n",
            "Epoch 118/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0033 - mae: 0.0610 - val_loss: 0.0033 - val_mae: 0.0633\n",
            "Epoch 119/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0563 - val_loss: 0.0031 - val_mae: 0.0615\n",
            "Epoch 120/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0034 - mae: 0.0597 - val_loss: 0.0032 - val_mae: 0.0626\n",
            "Epoch 121/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0027 - mae: 0.0546 - val_loss: 0.0033 - val_mae: 0.0638\n",
            "Epoch 122/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0567 - val_loss: 0.0031 - val_mae: 0.0605\n",
            "Epoch 123/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0030 - mae: 0.0579 - val_loss: 0.0031 - val_mae: 0.0610\n",
            "Epoch 124/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0029 - mae: 0.0571 - val_loss: 0.0031 - val_mae: 0.0610\n",
            "Epoch 125/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0026 - mae: 0.0546 - val_loss: 0.0030 - val_mae: 0.0598\n",
            "Epoch 126/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0030 - mae: 0.0566 - val_loss: 0.0030 - val_mae: 0.0600\n",
            "Epoch 127/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0026 - mae: 0.0547 - val_loss: 0.0029 - val_mae: 0.0584\n",
            "Epoch 128/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0574 - val_loss: 0.0028 - val_mae: 0.0580\n",
            "Epoch 129/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0027 - mae: 0.0561 - val_loss: 0.0028 - val_mae: 0.0576\n",
            "Epoch 130/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0535 - val_loss: 0.0029 - val_mae: 0.0590\n",
            "Epoch 131/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0024 - mae: 0.0521 - val_loss: 0.0028 - val_mae: 0.0579\n",
            "Epoch 132/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0565 - val_loss: 0.0027 - val_mae: 0.0564\n",
            "Epoch 133/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0558 - val_loss: 0.0028 - val_mae: 0.0583\n",
            "Epoch 134/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0029 - mae: 0.0554 - val_loss: 0.0029 - val_mae: 0.0587\n",
            "Epoch 135/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0026 - mae: 0.0531 - val_loss: 0.0028 - val_mae: 0.0577\n",
            "Epoch 136/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0032 - mae: 0.0587 - val_loss: 0.0029 - val_mae: 0.0591\n",
            "Epoch 137/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0027 - mae: 0.0542 - val_loss: 0.0027 - val_mae: 0.0566\n",
            "Epoch 138/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0030 - mae: 0.0578 - val_loss: 0.0027 - val_mae: 0.0563\n",
            "Epoch 139/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0555 - val_loss: 0.0027 - val_mae: 0.0562\n",
            "Epoch 140/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0540 - val_loss: 0.0026 - val_mae: 0.0559\n",
            "Epoch 141/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0030 - mae: 0.0577 - val_loss: 0.0027 - val_mae: 0.0567\n",
            "Epoch 142/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0029 - mae: 0.0552 - val_loss: 0.0027 - val_mae: 0.0566\n",
            "Epoch 143/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0029 - mae: 0.0550 - val_loss: 0.0026 - val_mae: 0.0558\n",
            "Epoch 144/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0027 - mae: 0.0543 - val_loss: 0.0026 - val_mae: 0.0559\n",
            "Epoch 145/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0032 - mae: 0.0586 - val_loss: 0.0026 - val_mae: 0.0560\n",
            "Epoch 146/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0029 - mae: 0.0542 - val_loss: 0.0027 - val_mae: 0.0561\n",
            "Epoch 147/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0025 - mae: 0.0547 - val_loss: 0.0025 - val_mae: 0.0542\n",
            "Epoch 148/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0026 - mae: 0.0531 - val_loss: 0.0026 - val_mae: 0.0558\n",
            "Epoch 149/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0031 - mae: 0.0585 - val_loss: 0.0026 - val_mae: 0.0554\n",
            "Epoch 150/150\n",
            "718/718 [==============================] - 2s 3ms/step - loss: 0.0028 - mae: 0.0554 - val_loss: 0.0026 - val_mae: 0.0556\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "id": "zfViRmNh6AiW",
        "outputId": "56d43f09-3969-4a34-a78a-1f11768e4463"
      },
      "source": [
        "mae = history.history['mae']\r\n",
        "val_mae = history.history['val_mae']\r\n",
        "loss = history.history['loss']\r\n",
        "val_loss = history.history['val_loss']\r\n",
        "\r\n",
        "epochs = range(len(mae))\r\n",
        "\r\n",
        "plt.plot(epochs, mae, 'r', label='Training MAE')\r\n",
        "plt.plot(epochs, val_mae, 'b', label='Validation MAE')\r\n",
        "plt.title('Training and validation MAE')\r\n",
        "plt.legend()\r\n",
        "plt.figure()\r\n",
        "\r\n",
        "plt.plot(epochs, loss, 'r', label='Training Loss')\r\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation Loss')\r\n",
        "plt.title('Training and validation loss')\r\n",
        "plt.legend()\r\n",
        "\r\n",
        "plt.show()"
      ],
      "execution_count": 435,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D46THA3X6odu",
        "outputId": "a38328ed-95d3-47e3-b5a0-a390c04af552"
      },
      "source": [
        "test_data = test_sets.data.copy()\r\n",
        "for i in range(len(test_data)-3):\r\n",
        "  topred = test_data[i+1:i+3,0]\r\n",
        "  topred = topred.reshape(1, 2, 1)\r\n",
        "  prediction = model.predict_generator(topred)\r\n",
        "  test_data[2+i] = prediction"
      ],
      "execution_count": 493,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1905: UserWarning: `Model.predict_generator` is deprecated and will be removed in a future version. Please use `Model.predict`, which supports generators.\n",
            "  warnings.warn('`Model.predict_generator` is deprecated and '\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f41KSk86Rkhq",
        "outputId": "75c8d624-cbb5-4b02-b871-f0d9f4657024"
      },
      "source": [
        "df_test['Pred'] = test_data"
      ],
      "execution_count": 478,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nladAnU_WWYP"
      },
      "source": [
        "rescale1 = minmaxscaler.inverse_transform(df_test[[\"AEP_MW\"]])\r\n",
        "rescale2 = minmaxscaler.inverse_transform(df_test[[\"Pred\"]])"
      ],
      "execution_count": 479,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qcEtNXldXKM6",
        "outputId": "d924abb8-97f1-42b8-cbde-df56d4764cbf"
      },
      "source": [
        "df_test[[\"AEP_MW\"]] = rescale1\r\n",
        "df_test[[\"Pred\"]] = rescale2"
      ],
      "execution_count": 480,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \"\"\"Entry point for launching an IPython kernel.\n",
            "/usr/local/lib/python3.6/dist-packages/pandas/core/indexing.py:1734: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  isetter(loc, value[:, i].tolist())\n",
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  \n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 338
        },
        "id": "r9nV_O8wSVIW",
        "outputId": "62ed699a-8799-400d-c666-210f5bd92624"
      },
      "source": [
        "df_test.plot(x='Datetime', y=[\"AEP_MW\",\"Pred\"], figsize=(15,5))\r\n",
        "plt.show()"
      ],
      "execution_count": 483,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}