DCP_LSTM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DCP-LSTM.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyN3w6qeB1BEEL3h0EHNIYa5",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/theAkhileshRai/DeepLearning/blob/master/DCP_LSTM.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZXGJtOxOn7CN",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import sklearn\n",
        "\n",
        "from numpy import loadtxt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.preprocessing import StandardScaler,OneHotEncoder    \n",
        "\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "pd.set_option('display.float_format', lambda x: '%.4f' % x)\n",
        "import seaborn as sns\n",
        "sns.set_context(\"paper\", font_scale=1.3)\n",
        "sns.set_style('white')\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "from time import time\n",
        "import matplotlib.ticker as tkr\n",
        "from scipy import stats\n",
        "from statsmodels.tsa.stattools import adfuller\n",
        "from sklearn import preprocessing\n",
        "from statsmodels.tsa.stattools import pacf\n",
        "%matplotlib inline\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "\n",
        "from keras.utils import to_categorical\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "pd.set_option('display.float_format', lambda x: '%.4f' % x)\n",
        "import seaborn as sns\n",
        "sns.set_context(\"paper\", font_scale=1.3)\n",
        "sns.set_style('white')\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "from time import time\n",
        "import matplotlib.ticker as tkr\n",
        "from scipy import stats\n",
        "from statsmodels.tsa.stattools import adfuller\n",
        "from sklearn import preprocessing\n",
        "from statsmodels.tsa.stattools import pacf\n",
        "%matplotlib inline\n",
        "import math\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense\n",
        "from keras.layers import LSTM\n",
        "from keras.layers import Dropout\n",
        "from keras.layers import *\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "from keras.callbacks import EarlyStopping\n",
        "from sklearn.preprocessing import StandardScaler"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sHGwBRAzkPO6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df = pd.read_csv('/content/DCP_leak_classification_train_data.csv',index_col=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y0va_VJjkaEQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 411
        },
        "outputId": "a4277483-415c-4b0e-972f-ead4c9206fdf"
      },
      "source": [
        "df.head()"
      ],
      "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>Unnamed: 0</th>\n",
              "      <th>Unnamed: 0.1</th>\n",
              "      <th>time</th>\n",
              "      <th>cyl</th>\n",
              "      <th>s_temp_crank</th>\n",
              "      <th>d_temp_crank</th>\n",
              "      <th>s_pressure_crank</th>\n",
              "      <th>d_pressure_crank</th>\n",
              "      <th>rpm_crank</th>\n",
              "      <th>s_temp_mean</th>\n",
              "      <th>d_temp_mean</th>\n",
              "      <th>s_pressure_mean</th>\n",
              "      <th>d_pressure_mean</th>\n",
              "      <th>rpm_mean</th>\n",
              "      <th>Asset</th>\n",
              "      <th>Binary</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2019-12-01 00:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>103.1136</td>\n",
              "      <td>216.5912</td>\n",
              "      <td>354.6428</td>\n",
              "      <td>836.7407</td>\n",
              "      <td>1357.9325</td>\n",
              "      <td>103.1136</td>\n",
              "      <td>218.0124</td>\n",
              "      <td>344.2387</td>\n",
              "      <td>820.9005</td>\n",
              "      <td>1357.9325</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2019-12-01 01:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>99.9650</td>\n",
              "      <td>214.1369</td>\n",
              "      <td>352.4248</td>\n",
              "      <td>837.7143</td>\n",
              "      <td>1358.2400</td>\n",
              "      <td>99.9650</td>\n",
              "      <td>215.2369</td>\n",
              "      <td>342.9925</td>\n",
              "      <td>822.1540</td>\n",
              "      <td>1358.2400</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2019-12-01 02:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>98.5032</td>\n",
              "      <td>213.4321</td>\n",
              "      <td>352.1486</td>\n",
              "      <td>842.5627</td>\n",
              "      <td>1398.2813</td>\n",
              "      <td>98.5032</td>\n",
              "      <td>214.1066</td>\n",
              "      <td>344.6910</td>\n",
              "      <td>829.0805</td>\n",
              "      <td>1398.2813</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2019-12-01 03:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>104.6202</td>\n",
              "      <td>220.9732</td>\n",
              "      <td>352.7063</td>\n",
              "      <td>848.7154</td>\n",
              "      <td>1397.8712</td>\n",
              "      <td>104.6202</td>\n",
              "      <td>221.6973</td>\n",
              "      <td>345.2832</td>\n",
              "      <td>835.5201</td>\n",
              "      <td>1397.8712</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2019-12-01 04:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>109.5678</td>\n",
              "      <td>227.3392</td>\n",
              "      <td>351.3930</td>\n",
              "      <td>851.2240</td>\n",
              "      <td>1397.3100</td>\n",
              "      <td>109.5678</td>\n",
              "      <td>228.1149</td>\n",
              "      <td>344.0607</td>\n",
              "      <td>838.4339</td>\n",
              "      <td>1397.3100</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Unnamed: 0  Unnamed: 0.1  ...            Asset  Binary\n",
              "0           0             0  ...  Ashland_Unit256  0.0000\n",
              "1           1             1  ...  Ashland_Unit256  0.0000\n",
              "2           2             2  ...  Ashland_Unit256  0.0000\n",
              "3           3             3  ...  Ashland_Unit256  0.0000\n",
              "4           4             4  ...  Ashland_Unit256  0.0000\n",
              "\n",
              "[5 rows x 16 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 121
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_WBE82GDkja6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e38c0f63-f85a-431b-abaf-0b90ccb7217b"
      },
      "source": [
        "len(df['Asset'].unique())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "19"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 122
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gZ24aQgtkx4N",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "05e04fac-a14a-40e5-e1c0-bb51e700146f"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(138498, 16)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c98EKWwEm2JL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df['time'] = pd.to_datetime(df['time'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-4rWYx9asf27",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#df['year'] = df['time'].apply(lambda x: x.year)\n",
        "df['quarter'] = df['time'].apply(lambda x: x.quarter)\n",
        "df['month'] = df['time'].apply(lambda x: x.month)\n",
        "df['day'] = df['time'].apply(lambda x: x.day)\n",
        "df['hour'] = df['time'].apply(lambda x: x.hour)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gJoiLOf4uD-5",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 411
        },
        "outputId": "1eafbb3d-663f-4618-ec4a-77a01900e40b"
      },
      "source": [
        "df.head()"
      ],
      "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>Unnamed: 0</th>\n",
              "      <th>Unnamed: 0.1</th>\n",
              "      <th>time</th>\n",
              "      <th>cyl</th>\n",
              "      <th>s_temp_crank</th>\n",
              "      <th>d_temp_crank</th>\n",
              "      <th>s_pressure_crank</th>\n",
              "      <th>d_pressure_crank</th>\n",
              "      <th>rpm_crank</th>\n",
              "      <th>s_temp_mean</th>\n",
              "      <th>d_temp_mean</th>\n",
              "      <th>s_pressure_mean</th>\n",
              "      <th>d_pressure_mean</th>\n",
              "      <th>rpm_mean</th>\n",
              "      <th>Asset</th>\n",
              "      <th>Binary</th>\n",
              "      <th>quarter</th>\n",
              "      <th>month</th>\n",
              "      <th>day</th>\n",
              "      <th>hour</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2019-12-01 00:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>103.1136</td>\n",
              "      <td>216.5912</td>\n",
              "      <td>354.6428</td>\n",
              "      <td>836.7407</td>\n",
              "      <td>1357.9325</td>\n",
              "      <td>103.1136</td>\n",
              "      <td>218.0124</td>\n",
              "      <td>344.2387</td>\n",
              "      <td>820.9005</td>\n",
              "      <td>1357.9325</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>4</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2019-12-01 01:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>99.9650</td>\n",
              "      <td>214.1369</td>\n",
              "      <td>352.4248</td>\n",
              "      <td>837.7143</td>\n",
              "      <td>1358.2400</td>\n",
              "      <td>99.9650</td>\n",
              "      <td>215.2369</td>\n",
              "      <td>342.9925</td>\n",
              "      <td>822.1540</td>\n",
              "      <td>1358.2400</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>4</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2019-12-01 02:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>98.5032</td>\n",
              "      <td>213.4321</td>\n",
              "      <td>352.1486</td>\n",
              "      <td>842.5627</td>\n",
              "      <td>1398.2813</td>\n",
              "      <td>98.5032</td>\n",
              "      <td>214.1066</td>\n",
              "      <td>344.6910</td>\n",
              "      <td>829.0805</td>\n",
              "      <td>1398.2813</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>4</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2019-12-01 03:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>104.6202</td>\n",
              "      <td>220.9732</td>\n",
              "      <td>352.7063</td>\n",
              "      <td>848.7154</td>\n",
              "      <td>1397.8712</td>\n",
              "      <td>104.6202</td>\n",
              "      <td>221.6973</td>\n",
              "      <td>345.2832</td>\n",
              "      <td>835.5201</td>\n",
              "      <td>1397.8712</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>4</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2019-12-01 04:30:00</td>\n",
              "      <td>1.0000</td>\n",
              "      <td>109.5678</td>\n",
              "      <td>227.3392</td>\n",
              "      <td>351.3930</td>\n",
              "      <td>851.2240</td>\n",
              "      <td>1397.3100</td>\n",
              "      <td>109.5678</td>\n",
              "      <td>228.1149</td>\n",
              "      <td>344.0607</td>\n",
              "      <td>838.4339</td>\n",
              "      <td>1397.3100</td>\n",
              "      <td>Ashland_Unit256</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>4</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Unnamed: 0  Unnamed: 0.1                time  ...  month  day  hour\n",
              "0           0             0 2019-12-01 00:30:00  ...     12    1     0\n",
              "1           1             1 2019-12-01 01:30:00  ...     12    1     1\n",
              "2           2             2 2019-12-01 02:30:00  ...     12    1     2\n",
              "3           3             3 2019-12-01 03:30:00  ...     12    1     3\n",
              "4           4             4 2019-12-01 04:30:00  ...     12    1     4\n",
              "\n",
              "[5 rows x 20 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 126
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oH1vIFtguMM-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "0a8c39d9-be2b-409f-f508-5a3c03c15f68"
      },
      "source": [
        "stat, p = stats.normaltest(df.Binary)\n",
        "print('Statistics=%.3f, p=%.3f' % (stat, p))\n",
        "alpha = 0.05\n",
        "if p > alpha:\n",
        "    print('Data looks Gaussian (fail to reject H0)')\n",
        "else:\n",
        "    print('Data does not look Gaussian (reject H0)')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Statistics=850048.453, p=0.000\n",
            "Data does not look Gaussian (reject H0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "56gY7PszvtiH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        },
        "outputId": "d9617011-e5a0-417e-b9e2-dc373a7ffbd8"
      },
      "source": [
        "sns.distplot(df.Binary);\n",
        "print( 'Kurtosis of normal distribution: {}'.format(stats.kurtosis(df.Binary)))\n",
        "print( 'Skewness of normal distribution: {}'.format(stats.skew(df.Binary)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Kurtosis of normal distribution: -1.4962378363397157\n",
            "Skewness of normal distribution: 0.7097620472103907\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": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_TV9sdgGzIHn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 385
        },
        "outputId": "86fc45fc-06b7-4559-b017-a20f6de8dbcf"
      },
      "source": [
        "df1=df.loc[:,['time','Binary']]\n",
        "df1.set_index('time',inplace=True)\n",
        "df1.plot(figsize=(12,5))\n",
        "plt.ylabel('Leak Classification')\n",
        "plt.legend().set_visible(False)\n",
        "plt.tight_layout()\n",
        "plt.title('Leak Time Series')\n",
        "sns.despine(top=True)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gi4-8ZvV33c8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3f6a6ba1-48d3-432e-ec74-d5f63f6e1a45"
      },
      "source": [
        "df['Binary'].shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(138498,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 130
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XRKLfbVo0iro",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "outputId": "6a47b668-7bec-415d-eaf7-15e9212be839"
      },
      "source": [
        "plt.figure(figsize=(14,5))\n",
        "plt.subplot(1,2,1)\n",
        "plt.subplots_adjust(wspace=0.2)\n",
        "sns.boxplot(x=\"month\", y=\"Binary\", data=df)\n",
        "plt.xlabel('monthly')\n",
        "plt.title('Box plot of Monthly Leak')\n",
        "sns.despine(left=True)\n",
        "plt.tight_layout()\n",
        "plt.subplot(1,2,2)\n",
        "sns.boxplot(x=\"quarter\", y=\"Binary\", data=df)\n",
        "plt.xlabel('quarter')\n",
        "plt.title('Box plot of Quarterly Leak')\n",
        "sns.despine(left=True)\n",
        "plt.tight_layout();"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n8UNLEz9wsyf",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "4e64f5c5-4d25-46b7-a1f4-17bd262b41e7"
      },
      "source": [
        "df.columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['Unnamed: 0', 'Unnamed: 0.1', 'time', 'cyl', 's_temp_crank',\n",
              "       'd_temp_crank', 's_pressure_crank', 'd_pressure_crank', 'rpm_crank',\n",
              "       's_temp_mean', 'd_temp_mean', 's_pressure_mean', 'd_pressure_mean',\n",
              "       'rpm_mean', 'Asset', 'Binary', 'quarter', 'month', 'day', 'hour'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 132
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mJ5rgm3Jy3Dg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_columns = [ 'quarter','month','day','hour','cyl', 's_temp_crank', 'd_temp_crank', 's_pressure_crank',\\\n",
        "       'd_pressure_crank', 'rpm_crank', 's_temp_mean', 'd_temp_mean',\\\n",
        "       's_pressure_mean', 'd_pressure_mean', 'rpm_mean']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y9q5u-vdy3nr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#time series lstm\n",
        "X = df[X_columns]\n",
        "y_true = df.Binary"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j-7RXYNC9iSx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "seed = 42\n",
        "test_size = 0.33\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y_true, test_size=test_size)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y8Dv4RHWMFAl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "# X_train = X_train.values\n",
        "# X_test = X_test.values\n",
        "X_train = np.reshape(X_train.values, (X_train.shape[0], 1, X_train.shape[1]))\n",
        "X_test = np.reshape(X_test.values, (X_test.shape[0], 1, X_test.shape[1]))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zcSzENiIJ3OW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2b5f0ffa-3232-437a-80cb-f32d00afbe83"
      },
      "source": [
        "y_test.size"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "45705"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 137
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eXJPf_fpy3dC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 632
        },
        "outputId": "94aa72f1-4b9a-41e6-dc87-4b207b7b4e09"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(LSTM(100, input_shape=(X_train.shape[1], X_train.shape[2])))\n",
        "#model.add(LSTM(100,input_shape=(X_train.shape[1], X_train.shape[2])))  # returns a sequence of vectors of dimension 30\n",
        "#model.add(LSTM(100,input_shape=(X_train.shape[1], X_train.shape[2])))  # return a single vector of dimension 30\n",
        "# model.add(Dense(50,activation='adam'))\n",
        "# model.add(Dense(25, activation='adam'))\n",
        "\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(1))\n",
        "model.compile(loss='binary_crossentropy', optimizer='adam',metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(X_train, y_train, epochs=20, batch_size=70, validation_data=(X_test, y_test), \n",
        "                    callbacks=[EarlyStopping(monitor='val_loss', patience=10)], verbose=1, shuffle=False)\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 92793 samples, validate on 45705 samples\n",
            "Epoch 1/20\n",
            "92793/92793 [==============================] - 6s 65us/step - loss: nan - accuracy: 3.6641e-04 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 2/20\n",
            "92793/92793 [==============================] - 5s 58us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 3/20\n",
            "92793/92793 [==============================] - 5s 59us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 4/20\n",
            "92793/92793 [==============================] - 5s 59us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 5/20\n",
            "92793/92793 [==============================] - 6s 61us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 6/20\n",
            "92793/92793 [==============================] - 6s 63us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 7/20\n",
            "92793/92793 [==============================] - 6s 60us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 8/20\n",
            "92793/92793 [==============================] - 5s 59us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 9/20\n",
            "92793/92793 [==============================] - 6s 59us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Epoch 10/20\n",
            "92793/92793 [==============================] - 5s 57us/step - loss: nan - accuracy: 0.0000e+00 - val_loss: nan - val_accuracy: 0.0000e+00\n",
            "Model: \"sequential_20\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "lstm_23 (LSTM)               (None, 100)               46400     \n",
            "_________________________________________________________________\n",
            "dropout_10 (Dropout)         (None, 100)               0         \n",
            "_________________________________________________________________\n",
            "dense_11 (Dense)             (None, 1)                 101       \n",
            "=================================================================\n",
            "Total params: 46,501\n",
            "Trainable params: 46,501\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WHQEkz3ly22w",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}