Build_With_AI_Hackathon.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Build With AI Hackathon.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "AmX_zKFUB4BC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%matplotlib inline \n",
        "import warnings  "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ThTtIxqeCPrU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "import sys \n",
        "import numpy as np \n",
        "from scipy.stats import randint\n",
        "import pandas as pd \n",
        "from matplotlib import pyplot\n",
        "import seaborn as sns \n",
        "\n",
        "\n",
        "\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from pandas import read_csv\n",
        "import math\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense\n",
        "from keras.layers import LSTM\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from keras.layers.core import Dense, Activation, Dropout\n",
        "import time #helper libraries"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gnw8ks9cCWHb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        },
        "outputId": "6bd7719c-0d7b-4601-a14e-b9565263c5d2"
      },
      "source": [
        "# load dataset\n",
        "dataset = pd.read_csv('final_dataset.csv', header=0, index_col=0)\n",
        "values = dataset.values\n",
        "# specify columns to plot\n",
        "groups = [0, 1, 2, 3, 5, 6, 7]\n",
        "i = 1\n",
        "# plot each column\n",
        "pyplot.figure()\n",
        "for group in groups:\n",
        "\tpyplot.subplot(len(groups), 1, i)\n",
        "\tpyplot.plot(values[:, group])\n",
        "\tpyplot.title(dataset.columns[group], y=1, loc='right')\n",
        "\ti += 1\n",
        "pyplot.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 7 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MT82b82bCi6C",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 390
        },
        "outputId": "41f380ed-5924-4fae-82b9-3d6707a93c75"
      },
      "source": [
        "dataset.head(10)"
      ],
      "execution_count": null,
      "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>0</th>\n",
              "      <th>0.1</th>\n",
              "      <th>1</th>\n",
              "      <th>0.2</th>\n",
              "      <th>0.3</th>\n",
              "      <th>0.4</th>\n",
              "      <th>0.5</th>\n",
              "      <th>1.1</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/22/20</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1/23/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/24/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/25/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/26/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/27/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/28/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/29/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/30/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1/31/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2/1/20</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         0  0.1  1  0.2  0.3  0.4  0.5  1.1\n",
              "1/22/20                                    \n",
              "1/23/20  0    0  0    0    0  0.0  0.0    1\n",
              "1/24/20  0    0  1    0    0  0.0  0.0    2\n",
              "1/25/20  0    0  0    0    0  0.0  0.0    2\n",
              "1/26/20  0    0  3    0    0  0.0  0.0    5\n",
              "1/27/20  0    0  0    0    0  0.0  0.0    5\n",
              "1/28/20  0    0  0    0    0  0.0  0.0    5\n",
              "1/29/20  0    0  0    0    0  0.0  0.0    5\n",
              "1/30/20  0    0  0    0    0  0.0  0.0    5\n",
              "1/31/20  0    0  2    0    0  0.0  0.0    7\n",
              "2/1/20   0    0  1    0    0  0.0  0.0    8"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9fDTitEVFC7s",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 170
        },
        "outputId": "43a334d7-776b-484b-91eb-2e6a54d519e1"
      },
      "source": [
        "dataset.isna().sum()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0      0\n",
              "0.1    0\n",
              "1      0\n",
              "0.2    0\n",
              "0.3    0\n",
              "0.4    0\n",
              "0.5    0\n",
              "1.1    0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 84
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IzNbQpayTNMt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b9c00f70-fceb-43d9-91cb-47e43445d9e5"
      },
      "source": [
        "dataset.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(185, 8)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eNyk_SouFW37",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "input_file=\"final_dataset.csv\"\n",
        "\n",
        "# convert an array of values into a dataset matrix\n",
        "def create_dataset(dataset, look_back=1):\n",
        "\tdataX, dataY = [], []\n",
        "\tfor i in range(len(dataset)-look_back-1):\n",
        "\t\ta = dataset[i:(i+look_back), 0]\n",
        "\t\tdataX.append(a)\n",
        "\t\tdataY.append(dataset[i + look_back, 0])\n",
        "\treturn np.array(dataX), np.array(dataY)\n",
        "\n",
        "# fix random seed for reproducibility\n",
        "# np.random.seed(5)\n",
        "\n",
        "# load the dataset\n",
        "df = read_csv(input_file, header=None, index_col=None, delimiter=',')\n",
        "\n",
        "# take cases columns\n",
        "all_y = df[8].values\n",
        "dataset=all_y.reshape(-1, 1)\n",
        "\n",
        "# normalize the dataset\n",
        "scaler = MinMaxScaler(feature_range=(0, 1))\n",
        "dataset = scaler.fit_transform(dataset)\n",
        "\n",
        "# split into train and test sets, 20% test data, 80% training data\n",
        "train_size = int(len(dataset) * 0.8)\n",
        "test_size = len(dataset) - train_size\n",
        "train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]\n",
        "\n",
        "# reshape into X=t and Y=t+1, timestep 12\n",
        "look_back = 12\n",
        "trainX, trainY = create_dataset(train, look_back)\n",
        "testX, testY = create_dataset(test, look_back)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oU5uJJq5U1ci",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "825b5aec-0198-445a-8ad8-8dd62fba7a1d"
      },
      "source": [
        "trainX.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(135, 12)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 87
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NpyRaLcyTcG4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "10b09a51-a603-4ae6-83bc-7d2dd2244863"
      },
      "source": [
        "\n",
        "# reshape input to be [samples, time steps, features]\n",
        "trainX = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))\n",
        "testX = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))\n",
        "\n",
        "# create and fit the LSTM network, optimizer=adam, 64 neurons, dropout 0.1\n",
        "model = Sequential()\n",
        "model.add(LSTM(64,activation='tanh', input_shape=(1, look_back)))\n",
        "model.add(Dropout(0.1))\n",
        "model.add(Dense(64,activation='relu'))\n",
        "\n",
        "model.add(Dense(1))\n",
        "model.compile(loss='mse', optimizer='adam')\n",
        "model.fit(trainX, trainY, epochs=500, validation_data=(testX,testY), batch_size=240, verbose=1)\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 135 samples, validate on 25 samples\n",
            "Epoch 1/500\n",
            "135/135 [==============================] - 0s 3ms/step - loss: 0.0566 - val_loss: 0.5553\n",
            "Epoch 2/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 0.0509 - val_loss: 0.4984\n",
            "Epoch 3/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 0.0435 - val_loss: 0.4480\n",
            "Epoch 4/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 0.0394 - val_loss: 0.3953\n",
            "Epoch 5/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 0.0333 - val_loss: 0.3481\n",
            "Epoch 6/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 0.0289 - val_loss: 0.3061\n",
            "Epoch 7/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 0.0238 - val_loss: 0.2682\n",
            "Epoch 8/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 0.0203 - val_loss: 0.2323\n",
            "Epoch 9/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0176 - val_loss: 0.1984\n",
            "Epoch 10/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0139 - val_loss: 0.1665\n",
            "Epoch 11/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 0.0110 - val_loss: 0.1369\n",
            "Epoch 12/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0090 - val_loss: 0.1095\n",
            "Epoch 13/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 0.0073 - val_loss: 0.0850\n",
            "Epoch 14/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 0.0059 - val_loss: 0.0639\n",
            "Epoch 15/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 0.0047 - val_loss: 0.0464\n",
            "Epoch 16/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 0.0044 - val_loss: 0.0322\n",
            "Epoch 17/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0041 - val_loss: 0.0211\n",
            "Epoch 18/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 0.0041 - val_loss: 0.0131\n",
            "Epoch 19/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 0.0041 - val_loss: 0.0077\n",
            "Epoch 20/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 0.0044 - val_loss: 0.0043\n",
            "Epoch 21/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 0.0046 - val_loss: 0.0024\n",
            "Epoch 22/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0045 - val_loss: 0.0013\n",
            "Epoch 23/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 0.0044 - val_loss: 8.3497e-04\n",
            "Epoch 24/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 0.0043 - val_loss: 6.3614e-04\n",
            "Epoch 25/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 0.0036 - val_loss: 6.6581e-04\n",
            "Epoch 26/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 0.0034 - val_loss: 8.8991e-04\n",
            "Epoch 27/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 0.0029 - val_loss: 0.0013\n",
            "Epoch 28/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 0.0025 - val_loss: 0.0021\n",
            "Epoch 29/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 0.0019 - val_loss: 0.0032\n",
            "Epoch 30/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 0.0016 - val_loss: 0.0045\n",
            "Epoch 31/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 0.0013 - val_loss: 0.0060\n",
            "Epoch 32/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 0.0011 - val_loss: 0.0076\n",
            "Epoch 33/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 9.3054e-04 - val_loss: 0.0090\n",
            "Epoch 34/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 8.5876e-04 - val_loss: 0.0101\n",
            "Epoch 35/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 8.4325e-04 - val_loss: 0.0108\n",
            "Epoch 36/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 8.1615e-04 - val_loss: 0.0111\n",
            "Epoch 37/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 8.9260e-04 - val_loss: 0.0109\n",
            "Epoch 38/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 7.4227e-04 - val_loss: 0.0103\n",
            "Epoch 39/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 40/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 8.9333e-04 - val_loss: 0.0080\n",
            "Epoch 41/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 7.0927e-04 - val_loss: 0.0064\n",
            "Epoch 42/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 6.6628e-04 - val_loss: 0.0049\n",
            "Epoch 43/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 6.9558e-04 - val_loss: 0.0035\n",
            "Epoch 44/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 5.8796e-04 - val_loss: 0.0022\n",
            "Epoch 45/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 4.8446e-04 - val_loss: 0.0013\n",
            "Epoch 46/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 4.3920e-04 - val_loss: 5.7540e-04\n",
            "Epoch 47/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 4.9662e-04 - val_loss: 1.8368e-04\n",
            "Epoch 48/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 4.8401e-04 - val_loss: 3.5080e-05\n",
            "Epoch 49/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 4.7311e-04 - val_loss: 5.7529e-05\n",
            "Epoch 50/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 6.2468e-04 - val_loss: 1.4817e-04\n",
            "Epoch 51/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 5.7911e-04 - val_loss: 2.3343e-04\n",
            "Epoch 52/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 5.7846e-04 - val_loss: 2.8182e-04\n",
            "Epoch 53/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 6.4347e-04 - val_loss: 2.5489e-04\n",
            "Epoch 54/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 5.4916e-04 - val_loss: 1.8554e-04\n",
            "Epoch 55/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 6.0764e-04 - val_loss: 9.3100e-05\n",
            "Epoch 56/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 5.3212e-04 - val_loss: 3.4607e-05\n",
            "Epoch 57/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 5.3373e-04 - val_loss: 4.5933e-05\n",
            "Epoch 58/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 4.4221e-04 - val_loss: 1.3714e-04\n",
            "Epoch 59/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 4.4277e-04 - val_loss: 2.9298e-04\n",
            "Epoch 60/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 4.9726e-04 - val_loss: 5.0493e-04\n",
            "Epoch 61/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 4.3235e-04 - val_loss: 7.6740e-04\n",
            "Epoch 62/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 4.5059e-04 - val_loss: 0.0010\n",
            "Epoch 63/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 4.5377e-04 - val_loss: 0.0013\n",
            "Epoch 64/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 4.8860e-04 - val_loss: 0.0015\n",
            "Epoch 65/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 5.3277e-04 - val_loss: 0.0016\n",
            "Epoch 66/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 3.8768e-04 - val_loss: 0.0017\n",
            "Epoch 67/500\n",
            "135/135 [==============================] - 0s 65us/step - loss: 4.0374e-04 - val_loss: 0.0017\n",
            "Epoch 68/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 4.3882e-04 - val_loss: 0.0016\n",
            "Epoch 69/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 4.8432e-04 - val_loss: 0.0015\n",
            "Epoch 70/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 3.9264e-04 - val_loss: 0.0014\n",
            "Epoch 71/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 4.4263e-04 - val_loss: 0.0012\n",
            "Epoch 72/500\n",
            "135/135 [==============================] - 0s 73us/step - loss: 3.8461e-04 - val_loss: 0.0011\n",
            "Epoch 73/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 4.4482e-04 - val_loss: 0.0011\n",
            "Epoch 74/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 4.5807e-04 - val_loss: 0.0011\n",
            "Epoch 75/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 3.7768e-04 - val_loss: 0.0011\n",
            "Epoch 76/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 4.6441e-04 - val_loss: 0.0012\n",
            "Epoch 77/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 3.3163e-04 - val_loss: 0.0013\n",
            "Epoch 78/500\n",
            "135/135 [==============================] - 0s 70us/step - loss: 4.1580e-04 - val_loss: 0.0014\n",
            "Epoch 79/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 3.9124e-04 - val_loss: 0.0016\n",
            "Epoch 80/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 3.1316e-04 - val_loss: 0.0018\n",
            "Epoch 81/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 4.3416e-04 - val_loss: 0.0020\n",
            "Epoch 82/500\n",
            "135/135 [==============================] - 0s 96us/step - loss: 4.2063e-04 - val_loss: 0.0022\n",
            "Epoch 83/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 3.8201e-04 - val_loss: 0.0023\n",
            "Epoch 84/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 3.1723e-04 - val_loss: 0.0025\n",
            "Epoch 85/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 3.7144e-04 - val_loss: 0.0026\n",
            "Epoch 86/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 3.1942e-04 - val_loss: 0.0026\n",
            "Epoch 87/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 3.7636e-04 - val_loss: 0.0027\n",
            "Epoch 88/500\n",
            "135/135 [==============================] - 0s 35us/step - loss: 2.9365e-04 - val_loss: 0.0027\n",
            "Epoch 89/500\n",
            "135/135 [==============================] - 0s 34us/step - loss: 4.1880e-04 - val_loss: 0.0027\n",
            "Epoch 90/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 4.0268e-04 - val_loss: 0.0026\n",
            "Epoch 91/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 3.2739e-04 - val_loss: 0.0025\n",
            "Epoch 92/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 3.6463e-04 - val_loss: 0.0025\n",
            "Epoch 93/500\n",
            "135/135 [==============================] - 0s 104us/step - loss: 3.9213e-04 - val_loss: 0.0025\n",
            "Epoch 94/500\n",
            "135/135 [==============================] - 0s 76us/step - loss: 3.8158e-04 - val_loss: 0.0026\n",
            "Epoch 95/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 4.0655e-04 - val_loss: 0.0028\n",
            "Epoch 96/500\n",
            "135/135 [==============================] - 0s 75us/step - loss: 3.2593e-04 - val_loss: 0.0029\n",
            "Epoch 97/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 3.7059e-04 - val_loss: 0.0031\n",
            "Epoch 98/500\n",
            "135/135 [==============================] - 0s 104us/step - loss: 3.1416e-04 - val_loss: 0.0032\n",
            "Epoch 99/500\n",
            "135/135 [==============================] - 0s 76us/step - loss: 3.5705e-04 - val_loss: 0.0032\n",
            "Epoch 100/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 3.4186e-04 - val_loss: 0.0032\n",
            "Epoch 101/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 3.6595e-04 - val_loss: 0.0031\n",
            "Epoch 102/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 3.9461e-04 - val_loss: 0.0030\n",
            "Epoch 103/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 2.9697e-04 - val_loss: 0.0029\n",
            "Epoch 104/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 2.8731e-04 - val_loss: 0.0029\n",
            "Epoch 105/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 3.9183e-04 - val_loss: 0.0029\n",
            "Epoch 106/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 3.6994e-04 - val_loss: 0.0030\n",
            "Epoch 107/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 3.6215e-04 - val_loss: 0.0030\n",
            "Epoch 108/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 3.4360e-04 - val_loss: 0.0031\n",
            "Epoch 109/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 3.4184e-04 - val_loss: 0.0031\n",
            "Epoch 110/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 3.3182e-04 - val_loss: 0.0032\n",
            "Epoch 111/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 3.3607e-04 - val_loss: 0.0033\n",
            "Epoch 112/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 3.1825e-04 - val_loss: 0.0035\n",
            "Epoch 113/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 3.3914e-04 - val_loss: 0.0037\n",
            "Epoch 114/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 3.2380e-04 - val_loss: 0.0039\n",
            "Epoch 115/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 3.8202e-04 - val_loss: 0.0041\n",
            "Epoch 116/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 2.2819e-04 - val_loss: 0.0044\n",
            "Epoch 117/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 2.4334e-04 - val_loss: 0.0046\n",
            "Epoch 118/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 3.4276e-04 - val_loss: 0.0046\n",
            "Epoch 119/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 3.4537e-04 - val_loss: 0.0046\n",
            "Epoch 120/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 2.8411e-04 - val_loss: 0.0047\n",
            "Epoch 121/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 2.8398e-04 - val_loss: 0.0046\n",
            "Epoch 122/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 2.0144e-04 - val_loss: 0.0047\n",
            "Epoch 123/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 2.8270e-04 - val_loss: 0.0048\n",
            "Epoch 124/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 2.7974e-04 - val_loss: 0.0048\n",
            "Epoch 125/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 2.6593e-04 - val_loss: 0.0048\n",
            "Epoch 126/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 2.2209e-04 - val_loss: 0.0050\n",
            "Epoch 127/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 2.8778e-04 - val_loss: 0.0051\n",
            "Epoch 128/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 2.2702e-04 - val_loss: 0.0053\n",
            "Epoch 129/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 3.0101e-04 - val_loss: 0.0054\n",
            "Epoch 130/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 2.3710e-04 - val_loss: 0.0055\n",
            "Epoch 131/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 2.8551e-04 - val_loss: 0.0056\n",
            "Epoch 132/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 2.4270e-04 - val_loss: 0.0056\n",
            "Epoch 133/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 2.9942e-04 - val_loss: 0.0056\n",
            "Epoch 134/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 2.5549e-04 - val_loss: 0.0055\n",
            "Epoch 135/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 2.3351e-04 - val_loss: 0.0055\n",
            "Epoch 136/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.8703e-04 - val_loss: 0.0056\n",
            "Epoch 137/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 2.3894e-04 - val_loss: 0.0057\n",
            "Epoch 138/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 3.3669e-04 - val_loss: 0.0058\n",
            "Epoch 139/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 3.0525e-04 - val_loss: 0.0059\n",
            "Epoch 140/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 2.8003e-04 - val_loss: 0.0060\n",
            "Epoch 141/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 2.5184e-04 - val_loss: 0.0062\n",
            "Epoch 142/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.7105e-04 - val_loss: 0.0063\n",
            "Epoch 143/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.9972e-04 - val_loss: 0.0065\n",
            "Epoch 144/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 2.5282e-04 - val_loss: 0.0066\n",
            "Epoch 145/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 2.0167e-04 - val_loss: 0.0067\n",
            "Epoch 146/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 2.8031e-04 - val_loss: 0.0067\n",
            "Epoch 147/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 2.8824e-04 - val_loss: 0.0065\n",
            "Epoch 148/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 3.5212e-04 - val_loss: 0.0063\n",
            "Epoch 149/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 2.5901e-04 - val_loss: 0.0062\n",
            "Epoch 150/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 1.8048e-04 - val_loss: 0.0060\n",
            "Epoch 151/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 2.5499e-04 - val_loss: 0.0060\n",
            "Epoch 152/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 2.9426e-04 - val_loss: 0.0058\n",
            "Epoch 153/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 2.8210e-04 - val_loss: 0.0056\n",
            "Epoch 154/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 2.3118e-04 - val_loss: 0.0056\n",
            "Epoch 155/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 2.1731e-04 - val_loss: 0.0058\n",
            "Epoch 156/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 2.3604e-04 - val_loss: 0.0059\n",
            "Epoch 157/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 2.0496e-04 - val_loss: 0.0059\n",
            "Epoch 158/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 1.8628e-04 - val_loss: 0.0061\n",
            "Epoch 159/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 2.4542e-04 - val_loss: 0.0062\n",
            "Epoch 160/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.9176e-04 - val_loss: 0.0064\n",
            "Epoch 161/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 2.9557e-04 - val_loss: 0.0066\n",
            "Epoch 162/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 2.8838e-04 - val_loss: 0.0066\n",
            "Epoch 163/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 2.2921e-04 - val_loss: 0.0067\n",
            "Epoch 164/500\n",
            "135/135 [==============================] - 0s 68us/step - loss: 2.0750e-04 - val_loss: 0.0068\n",
            "Epoch 165/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 2.7404e-04 - val_loss: 0.0068\n",
            "Epoch 166/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 2.2542e-04 - val_loss: 0.0070\n",
            "Epoch 167/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.2699e-04 - val_loss: 0.0071\n",
            "Epoch 168/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.1510e-04 - val_loss: 0.0072\n",
            "Epoch 169/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.5531e-04 - val_loss: 0.0075\n",
            "Epoch 170/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.9874e-04 - val_loss: 0.0078\n",
            "Epoch 171/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.8502e-04 - val_loss: 0.0079\n",
            "Epoch 172/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.7216e-04 - val_loss: 0.0082\n",
            "Epoch 173/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 1.5976e-04 - val_loss: 0.0082\n",
            "Epoch 174/500\n",
            "135/135 [==============================] - 0s 34us/step - loss: 2.2448e-04 - val_loss: 0.0082\n",
            "Epoch 175/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 2.2721e-04 - val_loss: 0.0083\n",
            "Epoch 176/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 2.1881e-04 - val_loss: 0.0082\n",
            "Epoch 177/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 1.8728e-04 - val_loss: 0.0083\n",
            "Epoch 178/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.7347e-04 - val_loss: 0.0084\n",
            "Epoch 179/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 2.3432e-04 - val_loss: 0.0083\n",
            "Epoch 180/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.9590e-04 - val_loss: 0.0084\n",
            "Epoch 181/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.6428e-04 - val_loss: 0.0084\n",
            "Epoch 182/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.6595e-04 - val_loss: 0.0083\n",
            "Epoch 183/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.7393e-04 - val_loss: 0.0083\n",
            "Epoch 184/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.9496e-04 - val_loss: 0.0083\n",
            "Epoch 185/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 1.4400e-04 - val_loss: 0.0081\n",
            "Epoch 186/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 2.9685e-04 - val_loss: 0.0080\n",
            "Epoch 187/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 2.2066e-04 - val_loss: 0.0079\n",
            "Epoch 188/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 2.1825e-04 - val_loss: 0.0077\n",
            "Epoch 189/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.5909e-04 - val_loss: 0.0077\n",
            "Epoch 190/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 1.8051e-04 - val_loss: 0.0078\n",
            "Epoch 191/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 2.1998e-04 - val_loss: 0.0078\n",
            "Epoch 192/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.7564e-04 - val_loss: 0.0079\n",
            "Epoch 193/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 2.2084e-04 - val_loss: 0.0082\n",
            "Epoch 194/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.5150e-04 - val_loss: 0.0086\n",
            "Epoch 195/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 2.2070e-04 - val_loss: 0.0089\n",
            "Epoch 196/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 2.6304e-04 - val_loss: 0.0090\n",
            "Epoch 197/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.4055e-04 - val_loss: 0.0092\n",
            "Epoch 198/500\n",
            "135/135 [==============================] - 0s 82us/step - loss: 1.6092e-04 - val_loss: 0.0093\n",
            "Epoch 199/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.3050e-04 - val_loss: 0.0095\n",
            "Epoch 200/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 2.6986e-04 - val_loss: 0.0094\n",
            "Epoch 201/500\n",
            "135/135 [==============================] - 0s 78us/step - loss: 2.0221e-04 - val_loss: 0.0091\n",
            "Epoch 202/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.3675e-04 - val_loss: 0.0088\n",
            "Epoch 203/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.9548e-04 - val_loss: 0.0086\n",
            "Epoch 204/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.7136e-04 - val_loss: 0.0084\n",
            "Epoch 205/500\n",
            "135/135 [==============================] - 0s 73us/step - loss: 1.8576e-04 - val_loss: 0.0083\n",
            "Epoch 206/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 2.1586e-04 - val_loss: 0.0083\n",
            "Epoch 207/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.7962e-04 - val_loss: 0.0085\n",
            "Epoch 208/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.9040e-04 - val_loss: 0.0085\n",
            "Epoch 209/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.6553e-04 - val_loss: 0.0087\n",
            "Epoch 210/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 1.7820e-04 - val_loss: 0.0087\n",
            "Epoch 211/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.2666e-04 - val_loss: 0.0086\n",
            "Epoch 212/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.4956e-04 - val_loss: 0.0085\n",
            "Epoch 213/500\n",
            "135/135 [==============================] - 0s 71us/step - loss: 2.2865e-04 - val_loss: 0.0083\n",
            "Epoch 214/500\n",
            "135/135 [==============================] - 0s 73us/step - loss: 1.6819e-04 - val_loss: 0.0082\n",
            "Epoch 215/500\n",
            "135/135 [==============================] - 0s 83us/step - loss: 2.1836e-04 - val_loss: 0.0082\n",
            "Epoch 216/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.9815e-04 - val_loss: 0.0082\n",
            "Epoch 217/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 1.7870e-04 - val_loss: 0.0083\n",
            "Epoch 218/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 2.8703e-04 - val_loss: 0.0084\n",
            "Epoch 219/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.1834e-04 - val_loss: 0.0085\n",
            "Epoch 220/500\n",
            "135/135 [==============================] - 0s 67us/step - loss: 1.4763e-04 - val_loss: 0.0089\n",
            "Epoch 221/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.9437e-04 - val_loss: 0.0092\n",
            "Epoch 222/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 1.4997e-04 - val_loss: 0.0094\n",
            "Epoch 223/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 1.5798e-04 - val_loss: 0.0095\n",
            "Epoch 224/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.3814e-04 - val_loss: 0.0095\n",
            "Epoch 225/500\n",
            "135/135 [==============================] - 0s 88us/step - loss: 1.5905e-04 - val_loss: 0.0096\n",
            "Epoch 226/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 2.2038e-04 - val_loss: 0.0094\n",
            "Epoch 227/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.3302e-04 - val_loss: 0.0092\n",
            "Epoch 228/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.8981e-04 - val_loss: 0.0092\n",
            "Epoch 229/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.2602e-04 - val_loss: 0.0093\n",
            "Epoch 230/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.4280e-04 - val_loss: 0.0094\n",
            "Epoch 231/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.7979e-04 - val_loss: 0.0093\n",
            "Epoch 232/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.3912e-04 - val_loss: 0.0091\n",
            "Epoch 233/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.1996e-04 - val_loss: 0.0091\n",
            "Epoch 234/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.6682e-04 - val_loss: 0.0091\n",
            "Epoch 235/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 1.8741e-04 - val_loss: 0.0092\n",
            "Epoch 236/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 2.2032e-04 - val_loss: 0.0092\n",
            "Epoch 237/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 1.6694e-04 - val_loss: 0.0094\n",
            "Epoch 238/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 2.4867e-04 - val_loss: 0.0095\n",
            "Epoch 239/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.8782e-04 - val_loss: 0.0094\n",
            "Epoch 240/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.4719e-04 - val_loss: 0.0092\n",
            "Epoch 241/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.5089e-04 - val_loss: 0.0091\n",
            "Epoch 242/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.5282e-04 - val_loss: 0.0092\n",
            "Epoch 243/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 1.2963e-04 - val_loss: 0.0092\n",
            "Epoch 244/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 2.1096e-04 - val_loss: 0.0091\n",
            "Epoch 245/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.5365e-04 - val_loss: 0.0091\n",
            "Epoch 246/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.2448e-04 - val_loss: 0.0094\n",
            "Epoch 247/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.5777e-04 - val_loss: 0.0098\n",
            "Epoch 248/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.9507e-04 - val_loss: 0.0100\n",
            "Epoch 249/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 1.4461e-04 - val_loss: 0.0100\n",
            "Epoch 250/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 2.2050e-04 - val_loss: 0.0099\n",
            "Epoch 251/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 2.0370e-04 - val_loss: 0.0095\n",
            "Epoch 252/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.7754e-04 - val_loss: 0.0093\n",
            "Epoch 253/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.6349e-04 - val_loss: 0.0091\n",
            "Epoch 254/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 1.5109e-04 - val_loss: 0.0088\n",
            "Epoch 255/500\n",
            "135/135 [==============================] - 0s 72us/step - loss: 1.3593e-04 - val_loss: 0.0087\n",
            "Epoch 256/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.4803e-04 - val_loss: 0.0088\n",
            "Epoch 257/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 2.3564e-04 - val_loss: 0.0088\n",
            "Epoch 258/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.0848e-04 - val_loss: 0.0090\n",
            "Epoch 259/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.5435e-04 - val_loss: 0.0095\n",
            "Epoch 260/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 1.8115e-04 - val_loss: 0.0100\n",
            "Epoch 261/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 2.2920e-04 - val_loss: 0.0100\n",
            "Epoch 262/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 1.3147e-04 - val_loss: 0.0099\n",
            "Epoch 263/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.3859e-04 - val_loss: 0.0098\n",
            "Epoch 264/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.4262e-04 - val_loss: 0.0096\n",
            "Epoch 265/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.2901e-04 - val_loss: 0.0093\n",
            "Epoch 266/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.4665e-04 - val_loss: 0.0091\n",
            "Epoch 267/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.3693e-04 - val_loss: 0.0091\n",
            "Epoch 268/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.4009e-04 - val_loss: 0.0090\n",
            "Epoch 269/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 1.5452e-04 - val_loss: 0.0090\n",
            "Epoch 270/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 1.3117e-04 - val_loss: 0.0091\n",
            "Epoch 271/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.3515e-04 - val_loss: 0.0092\n",
            "Epoch 272/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.4364e-04 - val_loss: 0.0093\n",
            "Epoch 273/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.3945e-04 - val_loss: 0.0091\n",
            "Epoch 274/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.3490e-04 - val_loss: 0.0092\n",
            "Epoch 275/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.0402e-04 - val_loss: 0.0092\n",
            "Epoch 276/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 2.0218e-04 - val_loss: 0.0090\n",
            "Epoch 277/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.6102e-04 - val_loss: 0.0088\n",
            "Epoch 278/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.3596e-04 - val_loss: 0.0086\n",
            "Epoch 279/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.4435e-04 - val_loss: 0.0083\n",
            "Epoch 280/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.7133e-04 - val_loss: 0.0082\n",
            "Epoch 281/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.1684e-04 - val_loss: 0.0081\n",
            "Epoch 282/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 1.1536e-04 - val_loss: 0.0082\n",
            "Epoch 283/500\n",
            "135/135 [==============================] - 0s 70us/step - loss: 1.4238e-04 - val_loss: 0.0082\n",
            "Epoch 284/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.1582e-04 - val_loss: 0.0083\n",
            "Epoch 285/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.4603e-04 - val_loss: 0.0084\n",
            "Epoch 286/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.4540e-04 - val_loss: 0.0083\n",
            "Epoch 287/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.1752e-04 - val_loss: 0.0083\n",
            "Epoch 288/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.6819e-04 - val_loss: 0.0083\n",
            "Epoch 289/500\n",
            "135/135 [==============================] - 0s 65us/step - loss: 1.2249e-04 - val_loss: 0.0084\n",
            "Epoch 290/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.1040e-04 - val_loss: 0.0084\n",
            "Epoch 291/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 1.4280e-04 - val_loss: 0.0087\n",
            "Epoch 292/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.1243e-04 - val_loss: 0.0087\n",
            "Epoch 293/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 1.1703e-04 - val_loss: 0.0089\n",
            "Epoch 294/500\n",
            "135/135 [==============================] - 0s 68us/step - loss: 1.3766e-04 - val_loss: 0.0091\n",
            "Epoch 295/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.0761e-04 - val_loss: 0.0093\n",
            "Epoch 296/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.2591e-04 - val_loss: 0.0096\n",
            "Epoch 297/500\n",
            "135/135 [==============================] - 0s 75us/step - loss: 1.3591e-04 - val_loss: 0.0097\n",
            "Epoch 298/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.2195e-04 - val_loss: 0.0100\n",
            "Epoch 299/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 1.9104e-04 - val_loss: 0.0100\n",
            "Epoch 300/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.4376e-04 - val_loss: 0.0098\n",
            "Epoch 301/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.2768e-04 - val_loss: 0.0094\n",
            "Epoch 302/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.4768e-04 - val_loss: 0.0091\n",
            "Epoch 303/500\n",
            "135/135 [==============================] - 0s 68us/step - loss: 1.3922e-04 - val_loss: 0.0091\n",
            "Epoch 304/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.4093e-04 - val_loss: 0.0094\n",
            "Epoch 305/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.5765e-04 - val_loss: 0.0094\n",
            "Epoch 306/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.6384e-04 - val_loss: 0.0094\n",
            "Epoch 307/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 2.0312e-04 - val_loss: 0.0090\n",
            "Epoch 308/500\n",
            "135/135 [==============================] - 0s 78us/step - loss: 1.2999e-04 - val_loss: 0.0085\n",
            "Epoch 309/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.8630e-04 - val_loss: 0.0082\n",
            "Epoch 310/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 2.1395e-04 - val_loss: 0.0081\n",
            "Epoch 311/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 1.8279e-04 - val_loss: 0.0081\n",
            "Epoch 312/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.4159e-04 - val_loss: 0.0084\n",
            "Epoch 313/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.3588e-04 - val_loss: 0.0090\n",
            "Epoch 314/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.6151e-04 - val_loss: 0.0093\n",
            "Epoch 315/500\n",
            "135/135 [==============================] - 0s 73us/step - loss: 9.3332e-05 - val_loss: 0.0094\n",
            "Epoch 316/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 9.0371e-05 - val_loss: 0.0096\n",
            "Epoch 317/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 1.1648e-04 - val_loss: 0.0098\n",
            "Epoch 318/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.4501e-04 - val_loss: 0.0096\n",
            "Epoch 319/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.2065e-04 - val_loss: 0.0092\n",
            "Epoch 320/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.2120e-04 - val_loss: 0.0091\n",
            "Epoch 321/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.6100e-04 - val_loss: 0.0091\n",
            "Epoch 322/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 9.5790e-05 - val_loss: 0.0090\n",
            "Epoch 323/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.2257e-04 - val_loss: 0.0091\n",
            "Epoch 324/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.9316e-04 - val_loss: 0.0089\n",
            "Epoch 325/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 1.1803e-04 - val_loss: 0.0086\n",
            "Epoch 326/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.2913e-04 - val_loss: 0.0083\n",
            "Epoch 327/500\n",
            "135/135 [==============================] - 0s 87us/step - loss: 1.0739e-04 - val_loss: 0.0084\n",
            "Epoch 328/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.3955e-04 - val_loss: 0.0084\n",
            "Epoch 329/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 1.0074e-04 - val_loss: 0.0086\n",
            "Epoch 330/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.1265e-04 - val_loss: 0.0091\n",
            "Epoch 331/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.8557e-04 - val_loss: 0.0095\n",
            "Epoch 332/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.5875e-04 - val_loss: 0.0097\n",
            "Epoch 333/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.5916e-04 - val_loss: 0.0097\n",
            "Epoch 334/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.0342e-04 - val_loss: 0.0098\n",
            "Epoch 335/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.0701e-04 - val_loss: 0.0099\n",
            "Epoch 336/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 2.2238e-04 - val_loss: 0.0097\n",
            "Epoch 337/500\n",
            "135/135 [==============================] - 0s 68us/step - loss: 9.1837e-05 - val_loss: 0.0095\n",
            "Epoch 338/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 8.0780e-05 - val_loss: 0.0093\n",
            "Epoch 339/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 8.8056e-05 - val_loss: 0.0092\n",
            "Epoch 340/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.0839e-04 - val_loss: 0.0093\n",
            "Epoch 341/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.3295e-04 - val_loss: 0.0094\n",
            "Epoch 342/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.4727e-04 - val_loss: 0.0091\n",
            "Epoch 343/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.2926e-04 - val_loss: 0.0088\n",
            "Epoch 344/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.3669e-04 - val_loss: 0.0088\n",
            "Epoch 345/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 9.2825e-05 - val_loss: 0.0089\n",
            "Epoch 346/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.6133e-04 - val_loss: 0.0087\n",
            "Epoch 347/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 1.1484e-04 - val_loss: 0.0089\n",
            "Epoch 348/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 1.0064e-04 - val_loss: 0.0089\n",
            "Epoch 349/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.0030e-04 - val_loss: 0.0089\n",
            "Epoch 350/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.0872e-04 - val_loss: 0.0090\n",
            "Epoch 351/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 1.2184e-04 - val_loss: 0.0092\n",
            "Epoch 352/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.3374e-04 - val_loss: 0.0096\n",
            "Epoch 353/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 6.7896e-05 - val_loss: 0.0100\n",
            "Epoch 354/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 8.3294e-05 - val_loss: 0.0104\n",
            "Epoch 355/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 1.4749e-04 - val_loss: 0.0105\n",
            "Epoch 356/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 8.5045e-05 - val_loss: 0.0108\n",
            "Epoch 357/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 1.4722e-04 - val_loss: 0.0108\n",
            "Epoch 358/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.3570e-04 - val_loss: 0.0105\n",
            "Epoch 359/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 6.9329e-05 - val_loss: 0.0103\n",
            "Epoch 360/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 1.0240e-04 - val_loss: 0.0099\n",
            "Epoch 361/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.2055e-04 - val_loss: 0.0095\n",
            "Epoch 362/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.4663e-04 - val_loss: 0.0091\n",
            "Epoch 363/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 1.0528e-04 - val_loss: 0.0089\n",
            "Epoch 364/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.3476e-04 - val_loss: 0.0090\n",
            "Epoch 365/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 9.4900e-05 - val_loss: 0.0091\n",
            "Epoch 366/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 1.3146e-04 - val_loss: 0.0092\n",
            "Epoch 367/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 1.1248e-04 - val_loss: 0.0093\n",
            "Epoch 368/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.0349e-04 - val_loss: 0.0092\n",
            "Epoch 369/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 9.8911e-05 - val_loss: 0.0092\n",
            "Epoch 370/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 7.4017e-05 - val_loss: 0.0093\n",
            "Epoch 371/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.0337e-04 - val_loss: 0.0094\n",
            "Epoch 372/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.3158e-04 - val_loss: 0.0095\n",
            "Epoch 373/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 8.3239e-05 - val_loss: 0.0095\n",
            "Epoch 374/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.3240e-04 - val_loss: 0.0094\n",
            "Epoch 375/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.0443e-04 - val_loss: 0.0094\n",
            "Epoch 376/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.5108e-04 - val_loss: 0.0092\n",
            "Epoch 377/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.1442e-04 - val_loss: 0.0091\n",
            "Epoch 378/500\n",
            "135/135 [==============================] - 0s 67us/step - loss: 8.2604e-05 - val_loss: 0.0092\n",
            "Epoch 379/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.0711e-04 - val_loss: 0.0089\n",
            "Epoch 380/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.2625e-04 - val_loss: 0.0087\n",
            "Epoch 381/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.4642e-04 - val_loss: 0.0080\n",
            "Epoch 382/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 1.0700e-04 - val_loss: 0.0074\n",
            "Epoch 383/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.2971e-04 - val_loss: 0.0073\n",
            "Epoch 384/500\n",
            "135/135 [==============================] - 0s 57us/step - loss: 1.1329e-04 - val_loss: 0.0077\n",
            "Epoch 385/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.3659e-04 - val_loss: 0.0084\n",
            "Epoch 386/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 9.4232e-05 - val_loss: 0.0092\n",
            "Epoch 387/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 1.0384e-04 - val_loss: 0.0099\n",
            "Epoch 388/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 1.2926e-04 - val_loss: 0.0101\n",
            "Epoch 389/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 1.0008e-04 - val_loss: 0.0100\n",
            "Epoch 390/500\n",
            "135/135 [==============================] - 0s 74us/step - loss: 6.0458e-05 - val_loss: 0.0099\n",
            "Epoch 391/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.4639e-04 - val_loss: 0.0096\n",
            "Epoch 392/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.5069e-04 - val_loss: 0.0094\n",
            "Epoch 393/500\n",
            "135/135 [==============================] - 0s 78us/step - loss: 1.5579e-04 - val_loss: 0.0091\n",
            "Epoch 394/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.2893e-04 - val_loss: 0.0092\n",
            "Epoch 395/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 9.8342e-05 - val_loss: 0.0092\n",
            "Epoch 396/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 1.4207e-04 - val_loss: 0.0092\n",
            "Epoch 397/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 9.4740e-05 - val_loss: 0.0092\n",
            "Epoch 398/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 9.1065e-05 - val_loss: 0.0090\n",
            "Epoch 399/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 9.3694e-05 - val_loss: 0.0089\n",
            "Epoch 400/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 9.8079e-05 - val_loss: 0.0085\n",
            "Epoch 401/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 1.0387e-04 - val_loss: 0.0082\n",
            "Epoch 402/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 1.1052e-04 - val_loss: 0.0078\n",
            "Epoch 403/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 8.7232e-05 - val_loss: 0.0082\n",
            "Epoch 404/500\n",
            "135/135 [==============================] - 0s 34us/step - loss: 1.3294e-04 - val_loss: 0.0084\n",
            "Epoch 405/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 6.3247e-05 - val_loss: 0.0088\n",
            "Epoch 406/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 8.5767e-05 - val_loss: 0.0092\n",
            "Epoch 407/500\n",
            "135/135 [==============================] - 0s 42us/step - loss: 1.3555e-04 - val_loss: 0.0095\n",
            "Epoch 408/500\n",
            "135/135 [==============================] - 0s 36us/step - loss: 1.2706e-04 - val_loss: 0.0096\n",
            "Epoch 409/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.0226e-04 - val_loss: 0.0097\n",
            "Epoch 410/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.1232e-04 - val_loss: 0.0094\n",
            "Epoch 411/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.0192e-04 - val_loss: 0.0089\n",
            "Epoch 412/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 9.6259e-05 - val_loss: 0.0087\n",
            "Epoch 413/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.1656e-04 - val_loss: 0.0083\n",
            "Epoch 414/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.4803e-04 - val_loss: 0.0080\n",
            "Epoch 415/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 8.8210e-05 - val_loss: 0.0078\n",
            "Epoch 416/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 1.1980e-04 - val_loss: 0.0080\n",
            "Epoch 417/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.4331e-04 - val_loss: 0.0085\n",
            "Epoch 418/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 9.4529e-05 - val_loss: 0.0090\n",
            "Epoch 419/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.2568e-04 - val_loss: 0.0094\n",
            "Epoch 420/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 1.0856e-04 - val_loss: 0.0097\n",
            "Epoch 421/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.1592e-04 - val_loss: 0.0098\n",
            "Epoch 422/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.0363e-04 - val_loss: 0.0096\n",
            "Epoch 423/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 8.9467e-05 - val_loss: 0.0094\n",
            "Epoch 424/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.1215e-04 - val_loss: 0.0092\n",
            "Epoch 425/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.0429e-04 - val_loss: 0.0093\n",
            "Epoch 426/500\n",
            "135/135 [==============================] - 0s 69us/step - loss: 1.7377e-04 - val_loss: 0.0097\n",
            "Epoch 427/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 1.0209e-04 - val_loss: 0.0099\n",
            "Epoch 428/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.0452e-04 - val_loss: 0.0099\n",
            "Epoch 429/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 1.0598e-04 - val_loss: 0.0100\n",
            "Epoch 430/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 6.7322e-05 - val_loss: 0.0104\n",
            "Epoch 431/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.5290e-04 - val_loss: 0.0104\n",
            "Epoch 432/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.2490e-04 - val_loss: 0.0098\n",
            "Epoch 433/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 8.4275e-05 - val_loss: 0.0092\n",
            "Epoch 434/500\n",
            "135/135 [==============================] - 0s 37us/step - loss: 1.0496e-04 - val_loss: 0.0087\n",
            "Epoch 435/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 6.6008e-05 - val_loss: 0.0086\n",
            "Epoch 436/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 9.8506e-05 - val_loss: 0.0084\n",
            "Epoch 437/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 9.1873e-05 - val_loss: 0.0085\n",
            "Epoch 438/500\n",
            "135/135 [==============================] - 0s 53us/step - loss: 7.0857e-05 - val_loss: 0.0087\n",
            "Epoch 439/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.0461e-04 - val_loss: 0.0088\n",
            "Epoch 440/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 1.2732e-04 - val_loss: 0.0088\n",
            "Epoch 441/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 9.5125e-05 - val_loss: 0.0085\n",
            "Epoch 442/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 7.8649e-05 - val_loss: 0.0083\n",
            "Epoch 443/500\n",
            "135/135 [==============================] - 0s 50us/step - loss: 1.4368e-04 - val_loss: 0.0079\n",
            "Epoch 444/500\n",
            "135/135 [==============================] - 0s 79us/step - loss: 1.2008e-04 - val_loss: 0.0080\n",
            "Epoch 445/500\n",
            "135/135 [==============================] - 0s 66us/step - loss: 1.1778e-04 - val_loss: 0.0081\n",
            "Epoch 446/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 8.2355e-05 - val_loss: 0.0081\n",
            "Epoch 447/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 9.1129e-05 - val_loss: 0.0085\n",
            "Epoch 448/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 5.3126e-05 - val_loss: 0.0090\n",
            "Epoch 449/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 1.2220e-04 - val_loss: 0.0091\n",
            "Epoch 450/500\n",
            "135/135 [==============================] - 0s 58us/step - loss: 1.0296e-04 - val_loss: 0.0090\n",
            "Epoch 451/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 8.4298e-05 - val_loss: 0.0086\n",
            "Epoch 452/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 8.6749e-05 - val_loss: 0.0083\n",
            "Epoch 453/500\n",
            "135/135 [==============================] - 0s 71us/step - loss: 8.0520e-05 - val_loss: 0.0081\n",
            "Epoch 454/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 9.4199e-05 - val_loss: 0.0084\n",
            "Epoch 455/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 6.9084e-05 - val_loss: 0.0088\n",
            "Epoch 456/500\n",
            "135/135 [==============================] - 0s 40us/step - loss: 9.2448e-05 - val_loss: 0.0090\n",
            "Epoch 457/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.0345e-04 - val_loss: 0.0089\n",
            "Epoch 458/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 7.6075e-05 - val_loss: 0.0089\n",
            "Epoch 459/500\n",
            "135/135 [==============================] - 0s 45us/step - loss: 6.7135e-05 - val_loss: 0.0090\n",
            "Epoch 460/500\n",
            "135/135 [==============================] - 0s 63us/step - loss: 9.1288e-05 - val_loss: 0.0090\n",
            "Epoch 461/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 9.0112e-05 - val_loss: 0.0092\n",
            "Epoch 462/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.4752e-04 - val_loss: 0.0091\n",
            "Epoch 463/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 7.2151e-05 - val_loss: 0.0091\n",
            "Epoch 464/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 9.2124e-05 - val_loss: 0.0088\n",
            "Epoch 465/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 7.5547e-05 - val_loss: 0.0087\n",
            "Epoch 466/500\n",
            "135/135 [==============================] - 0s 47us/step - loss: 1.0136e-04 - val_loss: 0.0088\n",
            "Epoch 467/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 1.0009e-04 - val_loss: 0.0089\n",
            "Epoch 468/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 1.2179e-04 - val_loss: 0.0085\n",
            "Epoch 469/500\n",
            "135/135 [==============================] - 0s 62us/step - loss: 8.6877e-05 - val_loss: 0.0083\n",
            "Epoch 470/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 9.3962e-05 - val_loss: 0.0083\n",
            "Epoch 471/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 9.6219e-05 - val_loss: 0.0084\n",
            "Epoch 472/500\n",
            "135/135 [==============================] - 0s 41us/step - loss: 8.4414e-05 - val_loss: 0.0083\n",
            "Epoch 473/500\n",
            "135/135 [==============================] - 0s 48us/step - loss: 7.4620e-05 - val_loss: 0.0084\n",
            "Epoch 474/500\n",
            "135/135 [==============================] - 0s 35us/step - loss: 1.0530e-04 - val_loss: 0.0086\n",
            "Epoch 475/500\n",
            "135/135 [==============================] - 0s 64us/step - loss: 7.4793e-05 - val_loss: 0.0091\n",
            "Epoch 476/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 1.0774e-04 - val_loss: 0.0090\n",
            "Epoch 477/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 7.9306e-05 - val_loss: 0.0086\n",
            "Epoch 478/500\n",
            "135/135 [==============================] - 0s 54us/step - loss: 6.5184e-05 - val_loss: 0.0085\n",
            "Epoch 479/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.0905e-04 - val_loss: 0.0083\n",
            "Epoch 480/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 8.5022e-05 - val_loss: 0.0081\n",
            "Epoch 481/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 1.0773e-04 - val_loss: 0.0080\n",
            "Epoch 482/500\n",
            "135/135 [==============================] - 0s 44us/step - loss: 9.9917e-05 - val_loss: 0.0079\n",
            "Epoch 483/500\n",
            "135/135 [==============================] - 0s 49us/step - loss: 9.5999e-05 - val_loss: 0.0079\n",
            "Epoch 484/500\n",
            "135/135 [==============================] - 0s 38us/step - loss: 7.9975e-05 - val_loss: 0.0080\n",
            "Epoch 485/500\n",
            "135/135 [==============================] - 0s 43us/step - loss: 1.0647e-04 - val_loss: 0.0080\n",
            "Epoch 486/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 9.6882e-05 - val_loss: 0.0082\n",
            "Epoch 487/500\n",
            "135/135 [==============================] - 0s 56us/step - loss: 1.0412e-04 - val_loss: 0.0086\n",
            "Epoch 488/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 9.7974e-05 - val_loss: 0.0087\n",
            "Epoch 489/500\n",
            "135/135 [==============================] - 0s 65us/step - loss: 7.7863e-05 - val_loss: 0.0091\n",
            "Epoch 490/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 1.0130e-04 - val_loss: 0.0089\n",
            "Epoch 491/500\n",
            "135/135 [==============================] - 0s 51us/step - loss: 7.7841e-05 - val_loss: 0.0088\n",
            "Epoch 492/500\n",
            "135/135 [==============================] - 0s 59us/step - loss: 7.9716e-05 - val_loss: 0.0088\n",
            "Epoch 493/500\n",
            "135/135 [==============================] - 0s 55us/step - loss: 8.6667e-05 - val_loss: 0.0088\n",
            "Epoch 494/500\n",
            "135/135 [==============================] - 0s 46us/step - loss: 7.8893e-05 - val_loss: 0.0089\n",
            "Epoch 495/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 1.0402e-04 - val_loss: 0.0090\n",
            "Epoch 496/500\n",
            "135/135 [==============================] - 0s 60us/step - loss: 6.1399e-05 - val_loss: 0.0093\n",
            "Epoch 497/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 8.2771e-05 - val_loss: 0.0092\n",
            "Epoch 498/500\n",
            "135/135 [==============================] - 0s 61us/step - loss: 1.0294e-04 - val_loss: 0.0090\n",
            "Epoch 499/500\n",
            "135/135 [==============================] - 0s 39us/step - loss: 8.6017e-05 - val_loss: 0.0085\n",
            "Epoch 500/500\n",
            "135/135 [==============================] - 0s 52us/step - loss: 9.1240e-05 - val_loss: 0.0078\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.callbacks.History at 0x7fd1f2b386a0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0gJtj2UIWxxV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "8be6d110-8a90-4432-e08b-0f30e120bb24"
      },
      "source": [
        "# make predictions\n",
        "trainPredict = model.predict(trainX)\n",
        "testPredict = model.predict(testX)\n",
        "\n",
        "# invert predictions\n",
        "trainPredict = scaler.inverse_transform(trainPredict)\n",
        "trainY = scaler.inverse_transform([trainY])\n",
        "testPredict = scaler.inverse_transform(testPredict)\n",
        "testY = scaler.inverse_transform([testY])\n",
        "\n",
        "# calculate root mean squared error\n",
        "trainScore = math.sqrt(mean_squared_error(trainY[0], trainPredict[:,0]))\n",
        "print('Train Score: %.2f RMSE' % (trainScore))\n",
        "testScore = math.sqrt(mean_squared_error(testY[0], testPredict[:,0]))\n",
        "print('Test Score: %.2f RMSE' % (testScore))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train Score: 22760.31 RMSE\n",
            "Test Score: 369558.98 RMSE\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3S247gzHXOyk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 755
        },
        "outputId": "e4dabfc2-231e-486f-acea-c8e1e6d7a3c9"
      },
      "source": [
        "\n",
        "# shift train predictions for plotting\n",
        "trainPredictPlot = np.empty_like(dataset)\n",
        "trainPredictPlot[:, :] = np.nan\n",
        "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict\n",
        "\n",
        "# shift test predictions for plotting\n",
        "testPredictPlot = np.empty_like(dataset)\n",
        "testPredictPlot[:, :] = np.nan\n",
        "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(dataset)-1, :] = testPredict\n",
        "\n",
        "# plot baseline and predictions\n",
        "plt.plot(scaler.inverse_transform(dataset))\n",
        "plt.plot(trainPredictPlot)\n",
        "print('testPrices:')\n",
        "testPrices=scaler.inverse_transform(dataset[test_size+look_back:])\n",
        "\n",
        "print('testPredictions:')\n",
        "print(testPredict)\n",
        "\n",
        "\n",
        "# plot the actual cases, prediction in test data=red line, actual cases=blue line\n",
        "plt.plot(testPredictPlot)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "testPrices:\n",
            "testPredictions:\n",
            "[[2435600.5]\n",
            " [2470476.5]\n",
            " [2506946.5]\n",
            " [2545143.8]\n",
            " [2585022. ]\n",
            " [2626152.5]\n",
            " [2668713.2]\n",
            " [2711290.5]\n",
            " [2755378.5]\n",
            " [2800036.8]\n",
            " [2846429.8]\n",
            " [2894761.8]\n",
            " [2944611.5]\n",
            " [2995397. ]\n",
            " [3046997.5]\n",
            " [3099688.5]\n",
            " [3152190.5]\n",
            " [3207962.5]\n",
            " [3264030. ]\n",
            " [3321538.8]\n",
            " [3379409.8]\n",
            " [3436835.8]\n",
            " [3493735.8]\n",
            " [3550642. ]\n",
            " [3607855. ]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7fd1f2827c18>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 90
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "71udUcYPV8Ll",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 721
        },
        "outputId": "67300b84-67d5-4341-f89f-26f394909669"
      },
      "source": [
        "print('testPredictions:')\n",
        "print(testPredict)\n",
        "\n",
        "\n",
        "# plot the actual cases, prediction in test data=red line, actual price=blue line\n",
        "plt.plot(testPredictPlot)\n",
        "plt.show()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "testPredictions:\n",
            "[[2435600.5]\n",
            " [2470476.5]\n",
            " [2506946.5]\n",
            " [2545143.8]\n",
            " [2585022. ]\n",
            " [2626152.5]\n",
            " [2668713.2]\n",
            " [2711290.5]\n",
            " [2755378.5]\n",
            " [2800036.8]\n",
            " [2846429.8]\n",
            " [2894761.8]\n",
            " [2944611.5]\n",
            " [2995397. ]\n",
            " [3046997.5]\n",
            " [3099688.5]\n",
            " [3152190.5]\n",
            " [3207962.5]\n",
            " [3264030. ]\n",
            " [3321538.8]\n",
            " [3379409.8]\n",
            " [3436835.8]\n",
            " [3493735.8]\n",
            " [3550642. ]\n",
            " [3607855. ]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B4yyvQtBWUR4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}