DCNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DCNN.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6JpxZ1-StcX6"
      },
      "source": [
        "##We are going to predict remaining useful lifetime (RUL) of a turbofan engine based on C-MAPSS data (Sensors and operating settings) using DCNN model (Deep Convolutionnal Neural Networks) "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ojwsYp9oGrKy"
      },
      "source": [
        "## Importing necessary libraries "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9nU_881Yk-a7"
      },
      "source": [
        "from matplotlib import rcParams\n",
        "import seaborn as sns\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import csv\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import accuracy_score as accuracy\n",
        "from sklearn.metrics import recall_score as recall\n",
        "from sklearn.metrics import precision_score as precision\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.model_selection import GroupShuffleSplit\n",
        "from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error\n",
        "from sklearn.cluster import KMeans\n",
        "from yellowbrick.cluster import KElbowVisualizer\n",
        "import tensorflow as tf \n",
        "import os\n",
        "import keras\n",
        "from keras.models import Sequential,load_model\n",
        "from keras.layers.recurrent import GRU\n",
        "from keras.layers.core import Dense, Activation, Dropout\n",
        "from numpy import newaxis\n",
        "from tensorflow.keras import optimizers\n",
        "from tensorflow.keras import Sequential\n",
        "from tensorflow.keras.layers import Dense, Dropout, Masking, TimeDistributed\n"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bn5Suh5xmZwW"
      },
      "source": [
        "##Exploratory Data Analysis "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OZ9fdj-2KHl8",
        "outputId": "fd0be9dc-afc7-433c-ecfb-3cac6638687f"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive/')"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive/\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-QBx9G0q2SJT"
      },
      "source": [
        "# Only raw data is provided without columns names, the lines bellow associate columns with their names \n",
        "x = ['sensor_' + str(i) for i in range(1, 22)]\n",
        "col_names = ['unit_number', 'time_cycles', 'op_setting_1', 'op_setting_2', 'op_setting_3'] + x"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JIc_r3OplPj9"
      },
      "source": [
        "train_01 = pd.read_csv('/content/drive/My Drive/Booz Allen/CMAPSSData/train_FD001.txt', delim_whitespace=True, header = None, names = col_names)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3EPqHzbB7JTH"
      },
      "source": [
        "test_01 = pd.read_csv('/content/drive/My Drive/Booz Allen/CMAPSSData/test_FD001.txt', delim_whitespace=True, header = None, names = col_names)"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nHa3iG99lix3"
      },
      "source": [
        "rul_01 = pd.read_csv('/content/drive/My Drive/Booz Allen/CMAPSSData/RUL_FD001.txt', delim_whitespace=True, header = None)"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 222
        },
        "id": "aSuNop93_GYd",
        "outputId": "b99eea91-8b1e-4420-8a3f-e512bf9acab1"
      },
      "source": [
        "train_01.head()"
      ],
      "execution_count": 7,
      "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>unit_number</th>\n",
              "      <th>time_cycles</th>\n",
              "      <th>op_setting_1</th>\n",
              "      <th>op_setting_2</th>\n",
              "      <th>op_setting_3</th>\n",
              "      <th>sensor_1</th>\n",
              "      <th>sensor_2</th>\n",
              "      <th>sensor_3</th>\n",
              "      <th>sensor_4</th>\n",
              "      <th>sensor_5</th>\n",
              "      <th>sensor_6</th>\n",
              "      <th>sensor_7</th>\n",
              "      <th>sensor_8</th>\n",
              "      <th>sensor_9</th>\n",
              "      <th>sensor_10</th>\n",
              "      <th>sensor_11</th>\n",
              "      <th>sensor_12</th>\n",
              "      <th>sensor_13</th>\n",
              "      <th>sensor_14</th>\n",
              "      <th>sensor_15</th>\n",
              "      <th>sensor_16</th>\n",
              "      <th>sensor_17</th>\n",
              "      <th>sensor_18</th>\n",
              "      <th>sensor_19</th>\n",
              "      <th>sensor_20</th>\n",
              "      <th>sensor_21</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.0007</td>\n",
              "      <td>-0.0004</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>641.82</td>\n",
              "      <td>1589.70</td>\n",
              "      <td>1400.60</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.36</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9046.19</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.47</td>\n",
              "      <td>521.66</td>\n",
              "      <td>2388.02</td>\n",
              "      <td>8138.62</td>\n",
              "      <td>8.4195</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.06</td>\n",
              "      <td>23.4190</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0.0019</td>\n",
              "      <td>-0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.15</td>\n",
              "      <td>1591.82</td>\n",
              "      <td>1403.14</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>553.75</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>9044.07</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.49</td>\n",
              "      <td>522.28</td>\n",
              "      <td>2388.07</td>\n",
              "      <td>8131.49</td>\n",
              "      <td>8.4318</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.00</td>\n",
              "      <td>23.4236</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>-0.0043</td>\n",
              "      <td>0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1587.99</td>\n",
              "      <td>1404.20</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.26</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>9052.94</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.27</td>\n",
              "      <td>522.42</td>\n",
              "      <td>2388.03</td>\n",
              "      <td>8133.23</td>\n",
              "      <td>8.4178</td>\n",
              "      <td>0.03</td>\n",
              "      <td>390</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.95</td>\n",
              "      <td>23.3442</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>0.0007</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1582.79</td>\n",
              "      <td>1401.87</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.45</td>\n",
              "      <td>2388.11</td>\n",
              "      <td>9049.48</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.13</td>\n",
              "      <td>522.86</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>8133.83</td>\n",
              "      <td>8.3682</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.88</td>\n",
              "      <td>23.3739</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>-0.0019</td>\n",
              "      <td>-0.0002</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.37</td>\n",
              "      <td>1582.85</td>\n",
              "      <td>1406.22</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.00</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9055.15</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.28</td>\n",
              "      <td>522.19</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>8133.80</td>\n",
              "      <td>8.4294</td>\n",
              "      <td>0.03</td>\n",
              "      <td>393</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.90</td>\n",
              "      <td>23.4044</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   unit_number  time_cycles  op_setting_1  ...  sensor_19  sensor_20  sensor_21\n",
              "0            1            1       -0.0007  ...      100.0      39.06    23.4190\n",
              "1            1            2        0.0019  ...      100.0      39.00    23.4236\n",
              "2            1            3       -0.0043  ...      100.0      38.95    23.3442\n",
              "3            1            4        0.0007  ...      100.0      38.88    23.3739\n",
              "4            1            5       -0.0019  ...      100.0      38.90    23.4044\n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 222
        },
        "id": "vck0MCpwTeTf",
        "outputId": "46c69971-6256-44d8-bc3d-561c9e8d0c40"
      },
      "source": [
        "train_01.head()"
      ],
      "execution_count": 8,
      "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>unit_number</th>\n",
              "      <th>time_cycles</th>\n",
              "      <th>op_setting_1</th>\n",
              "      <th>op_setting_2</th>\n",
              "      <th>op_setting_3</th>\n",
              "      <th>sensor_1</th>\n",
              "      <th>sensor_2</th>\n",
              "      <th>sensor_3</th>\n",
              "      <th>sensor_4</th>\n",
              "      <th>sensor_5</th>\n",
              "      <th>sensor_6</th>\n",
              "      <th>sensor_7</th>\n",
              "      <th>sensor_8</th>\n",
              "      <th>sensor_9</th>\n",
              "      <th>sensor_10</th>\n",
              "      <th>sensor_11</th>\n",
              "      <th>sensor_12</th>\n",
              "      <th>sensor_13</th>\n",
              "      <th>sensor_14</th>\n",
              "      <th>sensor_15</th>\n",
              "      <th>sensor_16</th>\n",
              "      <th>sensor_17</th>\n",
              "      <th>sensor_18</th>\n",
              "      <th>sensor_19</th>\n",
              "      <th>sensor_20</th>\n",
              "      <th>sensor_21</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.0007</td>\n",
              "      <td>-0.0004</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>641.82</td>\n",
              "      <td>1589.70</td>\n",
              "      <td>1400.60</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.36</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9046.19</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.47</td>\n",
              "      <td>521.66</td>\n",
              "      <td>2388.02</td>\n",
              "      <td>8138.62</td>\n",
              "      <td>8.4195</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.06</td>\n",
              "      <td>23.4190</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0.0019</td>\n",
              "      <td>-0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.15</td>\n",
              "      <td>1591.82</td>\n",
              "      <td>1403.14</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>553.75</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>9044.07</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.49</td>\n",
              "      <td>522.28</td>\n",
              "      <td>2388.07</td>\n",
              "      <td>8131.49</td>\n",
              "      <td>8.4318</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.00</td>\n",
              "      <td>23.4236</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>-0.0043</td>\n",
              "      <td>0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1587.99</td>\n",
              "      <td>1404.20</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.26</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>9052.94</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.27</td>\n",
              "      <td>522.42</td>\n",
              "      <td>2388.03</td>\n",
              "      <td>8133.23</td>\n",
              "      <td>8.4178</td>\n",
              "      <td>0.03</td>\n",
              "      <td>390</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.95</td>\n",
              "      <td>23.3442</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>0.0007</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1582.79</td>\n",
              "      <td>1401.87</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.45</td>\n",
              "      <td>2388.11</td>\n",
              "      <td>9049.48</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.13</td>\n",
              "      <td>522.86</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>8133.83</td>\n",
              "      <td>8.3682</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.88</td>\n",
              "      <td>23.3739</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>-0.0019</td>\n",
              "      <td>-0.0002</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.37</td>\n",
              "      <td>1582.85</td>\n",
              "      <td>1406.22</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.00</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9055.15</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.28</td>\n",
              "      <td>522.19</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>8133.80</td>\n",
              "      <td>8.4294</td>\n",
              "      <td>0.03</td>\n",
              "      <td>393</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.90</td>\n",
              "      <td>23.4044</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   unit_number  time_cycles  op_setting_1  ...  sensor_19  sensor_20  sensor_21\n",
              "0            1            1       -0.0007  ...      100.0      39.06    23.4190\n",
              "1            1            2        0.0019  ...      100.0      39.00    23.4236\n",
              "2            1            3       -0.0043  ...      100.0      38.95    23.3442\n",
              "3            1            4        0.0007  ...      100.0      38.88    23.3739\n",
              "4            1            5       -0.0019  ...      100.0      38.90    23.4044\n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "s3P-UAxlTgsl"
      },
      "source": [
        "##Data Preprocessing ==> the same preprocessing as in GRU model attached in this repo as well (please refer to that for more details)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KbKmXr31naGp"
      },
      "source": [
        "def add_operating_condition(df):\n",
        "    df_op_cond = df.copy()\n",
        "    \n",
        "    df_op_cond['op_setting_1'] = df_op_cond['op_setting_1'].round()\n",
        "    df_op_cond['op_setting_2'] = df_op_cond['op_setting_2'].round(decimals=2)\n",
        "    \n",
        "    # converting settings to string and concatanating makes the operating condition into a categorical variable\n",
        "    df_op_cond['op_cond'] = df_op_cond['op_setting_1'].astype(str) + '_' + \\\n",
        "                        df_op_cond['op_setting_2'].astype(str) + '_' + \\\n",
        "                        df_op_cond['op_setting_3'].astype(str)\n",
        "    \n",
        "    return df_op_cond\n",
        "\n",
        "def condition_scaler(df_train, df_test, sensor_names):\n",
        "    # apply operating condition specific scaling\n",
        "    scaler = StandardScaler()\n",
        "    for condition in df_train['op_cond'].unique():\n",
        "        scaler.fit(df_train.loc[df_train['op_cond']==condition, sensor_names])\n",
        "        df_train.loc[df_train['op_cond']==condition, sensor_names] = scaler.transform(df_train.loc[df_train['op_cond']==condition, sensor_names])\n",
        "        df_test.loc[df_test['op_cond']==condition, sensor_names] = scaler.transform(df_test.loc[df_test['op_cond']==condition, sensor_names])\n",
        "    return df_train, df_test"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SJXMSSd2xb2J"
      },
      "source": [
        "def exponential_smoothing(df, sensors, n_samples, alpha=0.4):\n",
        "    df = df.copy()\n",
        "    # first, take the exponential weighted mean\n",
        "    df[sensors] = df.groupby('unit_number')[sensors].apply(lambda x: x.ewm(alpha=alpha).mean())\n",
        "    # second, drop first n_samples of each unit_nr to reduce filter delay\n",
        "    def create_mask(data, samples):\n",
        "        result = np.ones_like(data)\n",
        "        result[0:samples] = 0\n",
        "        return result\n",
        "    \n",
        "    mask = df.groupby('unit_number')['unit_number'].transform(create_mask, samples=n_samples).astype(bool).reset_index(drop=True)\n",
        "    df = df[mask.iloc[:, 0]]\n",
        "    \n",
        "    return df"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qbC4ixOCxugj"
      },
      "source": [
        "gss = GroupShuffleSplit(n_splits=1, train_size=0.80, random_state=42)  # eventhough we set np and tf seeds, gss requires its own seed\n",
        "\n",
        "def train_val_group_split(X, y, gss, groups, print_groups=True):\n",
        "    \n",
        "    for idx_train, idx_val in gss.split(X, y, groups=groups):\n",
        "        if print_groups:\n",
        "            print('train_split_engines', train.iloc[idx_train]['unit_nr'].unique())\n",
        "            print('validate_split_engines', train.iloc[idx_val]['unit_nr'].unique(), '\\n')\n",
        "\n",
        "        X_train_split = X.iloc[idx_train].copy()\n",
        "        y_train_split = y.iloc[idx_train].copy()\n",
        "        X_val_split = X.iloc[idx_val].copy()\n",
        "        y_val_split = y.iloc[idx_val].copy()\n",
        "    return X_train_split, y_train_split, X_val_split, y_val_split"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KmG1mrhL00rq"
      },
      "source": [
        "#This function creates3D sequences of data , So far our data is 2D our model GRU needs 3D data \n",
        "def gen_train_data(df, sequence_length, columns):\n",
        "    data = df[columns].values\n",
        "    num_elements = data.shape[0]\n",
        "\n",
        "    # -1 and +1 because of Python indexing\n",
        "    for start, stop in zip(range(0, num_elements-(sequence_length-1)), range(sequence_length, num_elements+1)):\n",
        "        yield data[start:stop, :]\n"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_L3EG0tD2fUe"
      },
      "source": [
        "def gen_data_wrapper(df, sequence_length, columns, unit_nrs=np.array([])):\n",
        "    if unit_nrs.size <= 0:\n",
        "        unit_nrs = df['unit_number'].unique()\n",
        "        \n",
        "    data_gen = (list(gen_train_data(df[df['unit_number']==unit_nr], sequence_length, columns))\n",
        "               for unit_nr in unit_nrs)\n",
        "    data_array = np.concatenate(list(data_gen)).astype(np.float32)\n",
        "    return data_array"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jn7G5jbO25cj"
      },
      "source": [
        "def gen_labels(df, sequence_length, label):\n",
        "    data_matrix = df[label].values\n",
        "    num_elements = data_matrix.shape[0]\n",
        "\n",
        "    # -1 because I want to predict the rul of that last row in the sequence, not the next row\n",
        "    return data_matrix[sequence_length-1:num_elements, :]  \n",
        "\n",
        "def gen_label_wrapper(df, sequence_length, label, unit_nrs=np.array([])):\n",
        "    if unit_nrs.size <= 0:\n",
        "        unit_nrs = df['unit_number'].unique()\n",
        "        \n",
        "    label_gen = [gen_labels(df[df['unit_number']==unit_nr], sequence_length, label) \n",
        "                for unit_nr in unit_nrs]\n",
        "    label_array = np.concatenate(label_gen).astype(np.float32)\n",
        "    return label_array"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "twzUBF0K3wE4"
      },
      "source": [
        "def gen_test_data(df, sequence_length, columns, mask_value):\n",
        "    if df.shape[0] < sequence_length:\n",
        "        data_matrix = np.full(shape=(sequence_length, len(columns)), fill_value=mask_value) # pad\n",
        "        idx = data_matrix.shape[0] - df.shape[0]\n",
        "        data_matrix[idx:,:] = df[columns].values  # fill with available data\n",
        "    else:\n",
        "        data_matrix = df[columns].values\n",
        "        \n",
        "    # specifically yield the last possible sequence\n",
        "    stop = num_elements = data_matrix.shape[0]\n",
        "    start = stop - sequence_length\n",
        "    for i in list(range(1)):\n",
        "        yield data_matrix[start:stop, :]  "
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 222
        },
        "id": "psqZACeS3G89",
        "outputId": "5b148a18-69f4-4444-e3ea-4cb5c2d8e00c"
      },
      "source": [
        "def add_remaining_useful_life(df):\n",
        "    # Get the total number of cycles for each unit\n",
        "    grouped_by_unit = df.groupby(by=\"unit_number\")\n",
        "    max_cycle = grouped_by_unit[\"time_cycles\"].max()\n",
        "    \n",
        "    # Merge the max cycle back into the original frame\n",
        "    result_frame = df.merge(max_cycle.to_frame(name='max_cycle'), left_on='unit_number', right_index=True)\n",
        "    \n",
        "    # Calculate remaining useful life for each row\n",
        "    remaining_useful_life = result_frame[\"max_cycle\"] - result_frame[\"time_cycles\"]\n",
        "    result_frame[\"RUL\"] = remaining_useful_life\n",
        "    \n",
        "    # drop max_cycle as it's no longer needed\n",
        "    result_frame = result_frame.drop(\"max_cycle\", axis=1)\n",
        "    return result_frame\n",
        "\n",
        "train_01 = add_remaining_useful_life(train_01)\n",
        "train_01.head()"
      ],
      "execution_count": 16,
      "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>unit_number</th>\n",
              "      <th>time_cycles</th>\n",
              "      <th>op_setting_1</th>\n",
              "      <th>op_setting_2</th>\n",
              "      <th>op_setting_3</th>\n",
              "      <th>sensor_1</th>\n",
              "      <th>sensor_2</th>\n",
              "      <th>sensor_3</th>\n",
              "      <th>sensor_4</th>\n",
              "      <th>sensor_5</th>\n",
              "      <th>sensor_6</th>\n",
              "      <th>sensor_7</th>\n",
              "      <th>sensor_8</th>\n",
              "      <th>sensor_9</th>\n",
              "      <th>sensor_10</th>\n",
              "      <th>sensor_11</th>\n",
              "      <th>sensor_12</th>\n",
              "      <th>sensor_13</th>\n",
              "      <th>sensor_14</th>\n",
              "      <th>sensor_15</th>\n",
              "      <th>sensor_16</th>\n",
              "      <th>sensor_17</th>\n",
              "      <th>sensor_18</th>\n",
              "      <th>sensor_19</th>\n",
              "      <th>sensor_20</th>\n",
              "      <th>sensor_21</th>\n",
              "      <th>RUL</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>-0.0007</td>\n",
              "      <td>-0.0004</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>641.82</td>\n",
              "      <td>1589.70</td>\n",
              "      <td>1400.60</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.36</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9046.19</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.47</td>\n",
              "      <td>521.66</td>\n",
              "      <td>2388.02</td>\n",
              "      <td>8138.62</td>\n",
              "      <td>8.4195</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.06</td>\n",
              "      <td>23.4190</td>\n",
              "      <td>191</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0.0019</td>\n",
              "      <td>-0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.15</td>\n",
              "      <td>1591.82</td>\n",
              "      <td>1403.14</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>553.75</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>9044.07</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.49</td>\n",
              "      <td>522.28</td>\n",
              "      <td>2388.07</td>\n",
              "      <td>8131.49</td>\n",
              "      <td>8.4318</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>39.00</td>\n",
              "      <td>23.4236</td>\n",
              "      <td>190</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>-0.0043</td>\n",
              "      <td>0.0003</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1587.99</td>\n",
              "      <td>1404.20</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.26</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>9052.94</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.27</td>\n",
              "      <td>522.42</td>\n",
              "      <td>2388.03</td>\n",
              "      <td>8133.23</td>\n",
              "      <td>8.4178</td>\n",
              "      <td>0.03</td>\n",
              "      <td>390</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.95</td>\n",
              "      <td>23.3442</td>\n",
              "      <td>189</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>0.0007</td>\n",
              "      <td>0.0000</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.35</td>\n",
              "      <td>1582.79</td>\n",
              "      <td>1401.87</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.45</td>\n",
              "      <td>2388.11</td>\n",
              "      <td>9049.48</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.13</td>\n",
              "      <td>522.86</td>\n",
              "      <td>2388.08</td>\n",
              "      <td>8133.83</td>\n",
              "      <td>8.3682</td>\n",
              "      <td>0.03</td>\n",
              "      <td>392</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.88</td>\n",
              "      <td>23.3739</td>\n",
              "      <td>188</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>-0.0019</td>\n",
              "      <td>-0.0002</td>\n",
              "      <td>100.0</td>\n",
              "      <td>518.67</td>\n",
              "      <td>642.37</td>\n",
              "      <td>1582.85</td>\n",
              "      <td>1406.22</td>\n",
              "      <td>14.62</td>\n",
              "      <td>21.61</td>\n",
              "      <td>554.00</td>\n",
              "      <td>2388.06</td>\n",
              "      <td>9055.15</td>\n",
              "      <td>1.3</td>\n",
              "      <td>47.28</td>\n",
              "      <td>522.19</td>\n",
              "      <td>2388.04</td>\n",
              "      <td>8133.80</td>\n",
              "      <td>8.4294</td>\n",
              "      <td>0.03</td>\n",
              "      <td>393</td>\n",
              "      <td>2388</td>\n",
              "      <td>100.0</td>\n",
              "      <td>38.90</td>\n",
              "      <td>23.4044</td>\n",
              "      <td>187</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   unit_number  time_cycles  op_setting_1  ...  sensor_20  sensor_21  RUL\n",
              "0            1            1       -0.0007  ...      39.06    23.4190  191\n",
              "1            1            2        0.0019  ...      39.00    23.4236  190\n",
              "2            1            3       -0.0043  ...      38.95    23.3442  189\n",
              "3            1            4        0.0007  ...      38.88    23.3739  188\n",
              "4            1            5       -0.0019  ...      38.90    23.4044  187\n",
              "\n",
              "[5 rows x 27 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KEgTe125wvWD"
      },
      "source": [
        "sensors = ['sensor_' + str(i) for i in range(1, 22)]"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kJn-b_jK36Iz"
      },
      "source": [
        "sequence_length = 20\n",
        "train_01['RUL'].clip(upper=125, inplace=True)\n",
        "\n",
        "X_train_interim = add_operating_condition(train_01)\n",
        "X_test_interim = add_operating_condition(test_01)\n",
        "\n",
        "X_train_interim, X_test_interim = condition_scaler(X_train_interim, X_test_interim, sensors)\n",
        "\n",
        "# X_train_interim = exponential_smoothing(X_train_interim, sensors, 0, 0.4)\n",
        "# X_test_interim = exponential_smoothing(X_test_interim, sensors, 0, 0.4)\n",
        "\n",
        "# train-val split\n",
        "gss = GroupShuffleSplit(n_splits=1, train_size=0.80, random_state=42)\n",
        "for train_unit, val_unit in gss.split(X_train_interim['unit_number'].unique(), groups=X_train_interim['unit_number'].unique()):\n",
        "    train_unit = X_train_interim['unit_number'].unique()[train_unit]  # gss returns indexes and index starts at 1\n",
        "    val_unit = X_train_interim['unit_number'].unique()[val_unit]\n",
        "\n",
        "    train_split_array = gen_data_wrapper(X_train_interim, sequence_length, sensors, train_unit)\n",
        "    train_split_label = gen_label_wrapper(X_train_interim, sequence_length, ['RUL'], train_unit)\n",
        "    \n",
        "    val_split_array = gen_data_wrapper(X_train_interim, sequence_length, sensors, val_unit)\n",
        "    val_split_label = gen_label_wrapper(X_train_interim, sequence_length, ['RUL'], val_unit)\n",
        "\n",
        "# create sequences train, test \n",
        "train_array = gen_data_wrapper(X_train_interim, sequence_length, x)\n",
        "label_array = gen_label_wrapper(X_train_interim, sequence_length, ['RUL'])\n",
        "\n",
        "test_gen = (list(gen_test_data(X_test_interim[X_test_interim['unit_number']==unit_nr], sequence_length, x, -99.))\n",
        "           for unit_nr in X_test_interim['unit_number'].unique())\n",
        "test_array = np.concatenate(list(test_gen)).astype(np.float32)"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QJuAGjO_OV1w"
      },
      "source": [
        "#Modelling DCNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZMrfQo2QxkU8"
      },
      "source": [
        "nb_features = train_split_array.shape[2]\n",
        "nb_out = label_array.shape[1]"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ESqhUibikFyt"
      },
      "source": [
        " ### DCNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PyvJ4_U1q0xm"
      },
      "source": [
        "# Clear artifacts and set random seed to ensure repeatability\n",
        "keras.backend.clear_session()\n",
        "np.random.seed(42)\n",
        "#Create model\n",
        "def DCNN():\n",
        "    model = keras.models.Sequential()\n",
        "    model.add(keras.layers.Conv1D(32, 8, padding='same', activation='relu',\n",
        "                                  input_shape=(sequence_length,nb_features)))\n",
        "    model.add(keras.layers.Conv1D(64, 8, padding='same', activation='relu'))\n",
        "    model.add(keras.layers.Conv1D(64, 8, padding='same', activation='relu'))\n",
        "    model.add(keras.layers.Conv1D(32, 8, padding='same', activation='relu'))\n",
        "    model.add(keras.layers.GlobalMaxPooling1D())\n",
        "    model.add(keras.layers.Dropout(0.5))\n",
        "    model.add(keras.layers.Flatten())\n",
        "    model.add(keras.layers.Dense(100, activation='linear'))\n",
        "    model.add(keras.layers.Dropout(0.2))\n",
        "    model.add(keras.layers.Dense(units=nb_out))\n",
        "    model.compile(optimizer=tf.optimizers.Adam(learning_rate=0.0003),\n",
        "                  loss='mse',\n",
        "                  metrics=[tf.keras.metrics.RootMeanSquaredError()])\n",
        "    return model\n",
        "\n",
        "#Creat model checkpoint to save the best model performance\n",
        "model_checkpoint = keras.callbacks.ModelCheckpoint(\n",
        "    \"my_checkpoint.h5\", save_best_only=True)\n",
        "\n",
        "#specify ealy_callback to stop training model if there's no improvement after 10 epochs    \n",
        "early_stopping = keras.callbacks.EarlyStopping(patience=10)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_LOIM41Qr1tl",
        "outputId": "38dd2a2c-f2a0-4e35-fac1-a5c537642a82"
      },
      "source": [
        "# Check the model architecture\n",
        "New_DCNN=DCNN()\n",
        "New_DCNN.summary()"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv1d (Conv1D)             (None, 20, 32)            5408      \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 20, 64)            16448     \n",
            "                                                                 \n",
            " conv1d_2 (Conv1D)           (None, 20, 64)            32832     \n",
            "                                                                 \n",
            " conv1d_3 (Conv1D)           (None, 20, 32)            16416     \n",
            "                                                                 \n",
            " global_max_pooling1d (Globa  (None, 32)               0         \n",
            " lMaxPooling1D)                                                  \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 32)                0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 32)                0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 100)               3300      \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 100)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 1)                 101       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 74,505\n",
            "Trainable params: 74,505\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2i3Ap-Ti0R5i",
        "outputId": "056c23bd-4bbf-49ee-bb54-547a628744a5"
      },
      "source": [
        "# Train the model\n",
        "history = New_DCNN.fit(train_split_array,train_split_label, epochs=20, batch_size=200,\n",
        "                       validation_split=0.2,callbacks=[early_stopping, \n",
        "                        model_checkpoint])"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "61/61 [==============================] - 5s 67ms/step - loss: 5207.7080 - root_mean_squared_error: 72.1645 - val_loss: 1389.6442 - val_root_mean_squared_error: 37.2779\n",
            "Epoch 2/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 1572.0181 - root_mean_squared_error: 39.6487 - val_loss: 818.8135 - val_root_mean_squared_error: 28.6149\n",
            "Epoch 3/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 1073.5082 - root_mean_squared_error: 32.7644 - val_loss: 557.8494 - val_root_mean_squared_error: 23.6188\n",
            "Epoch 4/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 961.5441 - root_mean_squared_error: 31.0088 - val_loss: 505.1238 - val_root_mean_squared_error: 22.4750\n",
            "Epoch 5/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 962.2914 - root_mean_squared_error: 31.0208 - val_loss: 610.4301 - val_root_mean_squared_error: 24.7069\n",
            "Epoch 6/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 904.8553 - root_mean_squared_error: 30.0808 - val_loss: 455.5184 - val_root_mean_squared_error: 21.3429\n",
            "Epoch 7/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 868.7302 - root_mean_squared_error: 29.4742 - val_loss: 481.5510 - val_root_mean_squared_error: 21.9443\n",
            "Epoch 8/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 878.5479 - root_mean_squared_error: 29.6403 - val_loss: 390.0021 - val_root_mean_squared_error: 19.7485\n",
            "Epoch 9/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 878.6266 - root_mean_squared_error: 29.6416 - val_loss: 452.0604 - val_root_mean_squared_error: 21.2617\n",
            "Epoch 10/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 831.2032 - root_mean_squared_error: 28.8306 - val_loss: 380.0196 - val_root_mean_squared_error: 19.4941\n",
            "Epoch 11/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 842.4010 - root_mean_squared_error: 29.0241 - val_loss: 493.0606 - val_root_mean_squared_error: 22.2050\n",
            "Epoch 12/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 804.1345 - root_mean_squared_error: 28.3573 - val_loss: 487.6006 - val_root_mean_squared_error: 22.0817\n",
            "Epoch 13/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 817.5441 - root_mean_squared_error: 28.5927 - val_loss: 361.0990 - val_root_mean_squared_error: 19.0026\n",
            "Epoch 14/20\n",
            "61/61 [==============================] - 4s 63ms/step - loss: 804.3024 - root_mean_squared_error: 28.3602 - val_loss: 457.3760 - val_root_mean_squared_error: 21.3864\n",
            "Epoch 15/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 763.6533 - root_mean_squared_error: 27.6343 - val_loss: 383.1091 - val_root_mean_squared_error: 19.5732\n",
            "Epoch 16/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 783.3010 - root_mean_squared_error: 27.9875 - val_loss: 393.2238 - val_root_mean_squared_error: 19.8299\n",
            "Epoch 17/20\n",
            "61/61 [==============================] - 4s 62ms/step - loss: 760.5618 - root_mean_squared_error: 27.5783 - val_loss: 496.9915 - val_root_mean_squared_error: 22.2933\n",
            "Epoch 18/20\n",
            "61/61 [==============================] - 4s 64ms/step - loss: 717.0294 - root_mean_squared_error: 26.7774 - val_loss: 359.6613 - val_root_mean_squared_error: 18.9647\n",
            "Epoch 19/20\n",
            "61/61 [==============================] - 4s 64ms/step - loss: 718.7720 - root_mean_squared_error: 26.8099 - val_loss: 412.1805 - val_root_mean_squared_error: 20.3022\n",
            "Epoch 20/20\n",
            "61/61 [==============================] - 4s 64ms/step - loss: 723.5124 - root_mean_squared_error: 26.8982 - val_loss: 417.5438 - val_root_mean_squared_error: 20.4339\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EsdT8z0svUnn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 348
        },
        "outputId": "ffe858d8-11db-489b-b1fc-d3730c881ce5"
      },
      "source": [
        "# plot history\n",
        "def plot_loss(fit_history):\n",
        "    plt.figure(figsize=(13,5))\n",
        "    plt.plot(range(1, len(fit_history.history['loss'])+1), fit_history.history['loss'], label='train')\n",
        "    plt.plot(range(1, len(fit_history.history['val_loss'])+1), fit_history.history['val_loss'], label='validate')\n",
        "    plt.title('Deep Convolutionnal Neural Network (DCNN) model')\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "    plt.show()\n",
        "\n",
        "plot_loss(history)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 936x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YTJhPH-M6jvS"
      },
      "source": [
        "def evaluate(y_true, y_hat, label='test'):\n",
        "    mse = mean_squared_error(y_true, y_hat)\n",
        "    rmse = np.sqrt(mse)\n",
        "    mae = mean_absolute_error(y_true, y_hat)\n",
        "    variance = r2_score(y_true, y_hat)\n",
        "    print('{} set RMSE:{}, R2:{}, mae: {}'.format(label, rmse, variance, mae))"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BaKwdvVEkO4X",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e43130b6-f8e9-4919-a3ff-56bb67f69d60"
      },
      "source": [
        "# predict and evaluate\n",
        "y_hat_train = New_DCNN.predict(train_array)\n",
        "evaluate(label_array, y_hat_train, 'train')\n",
        "\n",
        "y_hat_test = New_DCNN.predict(test_array)\n",
        "evaluate(rul_01, y_hat_test)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "train set RMSE:18.14828109741211, R2:0.811894992926405, mae: 14.458889961242676\n",
            "test set RMSE:20.38985155458391, R2:0.7592487316830594, mae: 15.6485862159729\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kshxjZee8oiR"
      },
      "source": [
        "def plot_results(predicted_data, true_data): \n",
        "    fig = plt.figure(facecolor='white') \n",
        "    plt.figure(figsize=(13,5))\n",
        "    plt.plot(true_data, label='True Data') \n",
        "    plt.plot(predicted_data, label='Prediction') \n",
        "    plt.title('DCNN Model')\n",
        "    plt.legend() \n",
        "    plt.show() "
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 352
        },
        "id": "eUrLTi9W8p2n",
        "outputId": "420902af-3bde-44cb-8984-22be79206c04"
      },
      "source": [
        "plot_results(y_hat_test,rul_01)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 936x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qWWtEOHXvINq"
      },
      "source": [
        "##Conclusion: Results are acceptabl but I would go for GRU ;) "
      ]
    }
  ]
}