Comparision.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "Project_Final.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyNZaqcme3eh0/Kc5Fjb56Oy",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/maneeshsagar/Redshift-Estimation-of-low-to-high-Redshift-of-Galaxy-and-Quasars/blob/main/Comparision.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NIldWo_6aFTz",
        "outputId": "9525d842-eb73-450e-d75c-8e084aa54f7f"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/gdrive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1WYvxVFWZLPM"
      },
      "source": [
        "#This program is to develop a neural Network(ANN) regression model . redshift the label/target\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout\n",
        "from sklearn.preprocessing import  MinMaxScaler\n",
        "\n",
        "import xgboost as xg \n",
        "from sklearn.model_selection import train_test_split \n",
        "from sklearn.metrics import mean_squared_error as MSE "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hPoJqQysZZkJ"
      },
      "source": [
        "dataset = pd.read_csv(\"/content/gdrive/MyDrive/Dataset/dr16_train.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YStK8hOyTLEP",
        "outputId": "73170bb8-8ced-46d7-d00e-48d1bb4afee5"
      },
      "source": [
        "dataset.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3651936, 28)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 426
        },
        "id": "DKfuQwbyTYiX",
        "outputId": "af63d908-02c6-447a-d502-453df2eca3ab"
      },
      "source": [
        "dataset.loc[dataset.redshift<=1.00]"
      ],
      "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>objid</th>\n",
              "      <th>modelMag_u</th>\n",
              "      <th>modelMag_g</th>\n",
              "      <th>modelMag_r</th>\n",
              "      <th>modelMag_i</th>\n",
              "      <th>modelMag_z</th>\n",
              "      <th>fiberMag_u</th>\n",
              "      <th>fiberMag_g</th>\n",
              "      <th>fiberMag_r</th>\n",
              "      <th>fiberMag_i</th>\n",
              "      <th>fiberMag_z</th>\n",
              "      <th>petroR50_r</th>\n",
              "      <th>petroR90_r</th>\n",
              "      <th>petroR50_z</th>\n",
              "      <th>petroR90_z</th>\n",
              "      <th>r</th>\n",
              "      <th>i</th>\n",
              "      <th>z</th>\n",
              "      <th>redshift</th>\n",
              "      <th>zerr</th>\n",
              "      <th>mmug</th>\n",
              "      <th>mmgr</th>\n",
              "      <th>mmri</th>\n",
              "      <th>mmiz</th>\n",
              "      <th>mfug</th>\n",
              "      <th>mfgr</th>\n",
              "      <th>mfri</th>\n",
              "      <th>mfiz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>23.74654</td>\n",
              "      <td>23.19651</td>\n",
              "      <td>22.01303</td>\n",
              "      <td>20.80744</td>\n",
              "      <td>19.91660</td>\n",
              "      <td>24.60136</td>\n",
              "      <td>23.64122</td>\n",
              "      <td>22.48040</td>\n",
              "      <td>21.31004</td>\n",
              "      <td>20.34262</td>\n",
              "      <td>0.645596</td>\n",
              "      <td>1.295767</td>\n",
              "      <td>0.923503</td>\n",
              "      <td>2.043880</td>\n",
              "      <td>22.01303</td>\n",
              "      <td>20.80744</td>\n",
              "      <td>19.91660</td>\n",
              "      <td>0.814588</td>\n",
              "      <td>0.000212</td>\n",
              "      <td>23.471525</td>\n",
              "      <td>22.604770</td>\n",
              "      <td>21.410235</td>\n",
              "      <td>20.362020</td>\n",
              "      <td>24.121290</td>\n",
              "      <td>23.060810</td>\n",
              "      <td>21.895220</td>\n",
              "      <td>20.826330</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>22.18754</td>\n",
              "      <td>20.26853</td>\n",
              "      <td>18.66873</td>\n",
              "      <td>17.98356</td>\n",
              "      <td>17.62037</td>\n",
              "      <td>23.20970</td>\n",
              "      <td>21.22479</td>\n",
              "      <td>19.63727</td>\n",
              "      <td>18.96273</td>\n",
              "      <td>18.58847</td>\n",
              "      <td>1.641699</td>\n",
              "      <td>5.231110</td>\n",
              "      <td>1.148053</td>\n",
              "      <td>3.031853</td>\n",
              "      <td>18.66873</td>\n",
              "      <td>17.98356</td>\n",
              "      <td>17.62037</td>\n",
              "      <td>0.414405</td>\n",
              "      <td>0.000028</td>\n",
              "      <td>21.228035</td>\n",
              "      <td>19.468630</td>\n",
              "      <td>18.326145</td>\n",
              "      <td>17.801965</td>\n",
              "      <td>22.217245</td>\n",
              "      <td>20.431030</td>\n",
              "      <td>19.300000</td>\n",
              "      <td>18.775600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>19.85871</td>\n",
              "      <td>19.78292</td>\n",
              "      <td>19.52230</td>\n",
              "      <td>19.55171</td>\n",
              "      <td>19.33732</td>\n",
              "      <td>20.18514</td>\n",
              "      <td>20.09453</td>\n",
              "      <td>19.86055</td>\n",
              "      <td>19.90843</td>\n",
              "      <td>19.63532</td>\n",
              "      <td>0.611777</td>\n",
              "      <td>1.311494</td>\n",
              "      <td>0.584768</td>\n",
              "      <td>1.259439</td>\n",
              "      <td>19.52230</td>\n",
              "      <td>19.55171</td>\n",
              "      <td>19.33732</td>\n",
              "      <td>0.949956</td>\n",
              "      <td>0.000157</td>\n",
              "      <td>19.820815</td>\n",
              "      <td>19.652610</td>\n",
              "      <td>19.537005</td>\n",
              "      <td>19.444515</td>\n",
              "      <td>20.139835</td>\n",
              "      <td>19.977540</td>\n",
              "      <td>19.884490</td>\n",
              "      <td>19.771875</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>20.89670</td>\n",
              "      <td>20.47772</td>\n",
              "      <td>20.32096</td>\n",
              "      <td>20.23001</td>\n",
              "      <td>19.98116</td>\n",
              "      <td>21.25896</td>\n",
              "      <td>20.76655</td>\n",
              "      <td>20.65870</td>\n",
              "      <td>20.54609</td>\n",
              "      <td>20.17305</td>\n",
              "      <td>0.575010</td>\n",
              "      <td>1.252485</td>\n",
              "      <td>0.655031</td>\n",
              "      <td>1.284499</td>\n",
              "      <td>20.32096</td>\n",
              "      <td>20.23001</td>\n",
              "      <td>19.98116</td>\n",
              "      <td>0.703552</td>\n",
              "      <td>0.000085</td>\n",
              "      <td>20.687210</td>\n",
              "      <td>20.399340</td>\n",
              "      <td>20.275485</td>\n",
              "      <td>20.105585</td>\n",
              "      <td>21.012755</td>\n",
              "      <td>20.712625</td>\n",
              "      <td>20.602395</td>\n",
              "      <td>20.359570</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>20.44421</td>\n",
              "      <td>20.43502</td>\n",
              "      <td>20.29807</td>\n",
              "      <td>20.28478</td>\n",
              "      <td>19.87277</td>\n",
              "      <td>20.77106</td>\n",
              "      <td>20.74594</td>\n",
              "      <td>20.66329</td>\n",
              "      <td>20.64881</td>\n",
              "      <td>20.01088</td>\n",
              "      <td>0.663572</td>\n",
              "      <td>1.540812</td>\n",
              "      <td>0.928312</td>\n",
              "      <td>2.238584</td>\n",
              "      <td>20.29807</td>\n",
              "      <td>20.28478</td>\n",
              "      <td>19.87277</td>\n",
              "      <td>0.918946</td>\n",
              "      <td>0.000792</td>\n",
              "      <td>20.439615</td>\n",
              "      <td>20.366545</td>\n",
              "      <td>20.291425</td>\n",
              "      <td>20.078775</td>\n",
              "      <td>20.758500</td>\n",
              "      <td>20.704615</td>\n",
              "      <td>20.656050</td>\n",
              "      <td>20.329845</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3651918</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>23.55071</td>\n",
              "      <td>22.16176</td>\n",
              "      <td>21.42934</td>\n",
              "      <td>20.33605</td>\n",
              "      <td>19.45292</td>\n",
              "      <td>23.36273</td>\n",
              "      <td>22.77520</td>\n",
              "      <td>22.04431</td>\n",
              "      <td>20.97781</td>\n",
              "      <td>20.16630</td>\n",
              "      <td>0.902360</td>\n",
              "      <td>1.541793</td>\n",
              "      <td>1.343559</td>\n",
              "      <td>3.721590</td>\n",
              "      <td>21.42934</td>\n",
              "      <td>20.33605</td>\n",
              "      <td>19.45292</td>\n",
              "      <td>0.821233</td>\n",
              "      <td>0.000099</td>\n",
              "      <td>22.856235</td>\n",
              "      <td>21.795550</td>\n",
              "      <td>20.882695</td>\n",
              "      <td>19.894485</td>\n",
              "      <td>23.068965</td>\n",
              "      <td>22.409755</td>\n",
              "      <td>21.511060</td>\n",
              "      <td>20.572055</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3651919</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>21.86682</td>\n",
              "      <td>21.68359</td>\n",
              "      <td>21.37104</td>\n",
              "      <td>20.90272</td>\n",
              "      <td>20.47265</td>\n",
              "      <td>22.26036</td>\n",
              "      <td>22.02081</td>\n",
              "      <td>21.73457</td>\n",
              "      <td>21.26240</td>\n",
              "      <td>20.82135</td>\n",
              "      <td>0.718134</td>\n",
              "      <td>2.003701</td>\n",
              "      <td>0.557928</td>\n",
              "      <td>1.077190</td>\n",
              "      <td>21.37104</td>\n",
              "      <td>20.90272</td>\n",
              "      <td>20.47265</td>\n",
              "      <td>0.758018</td>\n",
              "      <td>0.000141</td>\n",
              "      <td>21.775205</td>\n",
              "      <td>21.527315</td>\n",
              "      <td>21.136880</td>\n",
              "      <td>20.687685</td>\n",
              "      <td>22.140585</td>\n",
              "      <td>21.877690</td>\n",
              "      <td>21.498485</td>\n",
              "      <td>21.041875</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3651923</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>22.31663</td>\n",
              "      <td>21.52032</td>\n",
              "      <td>21.18740</td>\n",
              "      <td>20.69922</td>\n",
              "      <td>20.88031</td>\n",
              "      <td>22.85945</td>\n",
              "      <td>21.85250</td>\n",
              "      <td>21.58473</td>\n",
              "      <td>21.05252</td>\n",
              "      <td>21.36186</td>\n",
              "      <td>0.579885</td>\n",
              "      <td>1.189824</td>\n",
              "      <td>0.323276</td>\n",
              "      <td>0.451393</td>\n",
              "      <td>21.18740</td>\n",
              "      <td>20.69922</td>\n",
              "      <td>20.88031</td>\n",
              "      <td>0.872583</td>\n",
              "      <td>0.000315</td>\n",
              "      <td>21.918475</td>\n",
              "      <td>21.353860</td>\n",
              "      <td>20.943310</td>\n",
              "      <td>20.789765</td>\n",
              "      <td>22.355975</td>\n",
              "      <td>21.718615</td>\n",
              "      <td>21.318625</td>\n",
              "      <td>21.207190</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3651924</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>26.08879</td>\n",
              "      <td>21.83270</td>\n",
              "      <td>21.34898</td>\n",
              "      <td>21.18016</td>\n",
              "      <td>21.19683</td>\n",
              "      <td>24.59543</td>\n",
              "      <td>22.18032</td>\n",
              "      <td>21.70586</td>\n",
              "      <td>21.61914</td>\n",
              "      <td>21.71572</td>\n",
              "      <td>0.763872</td>\n",
              "      <td>2.007011</td>\n",
              "      <td>0.652313</td>\n",
              "      <td>2.680450</td>\n",
              "      <td>21.34898</td>\n",
              "      <td>21.18016</td>\n",
              "      <td>21.19683</td>\n",
              "      <td>0.430104</td>\n",
              "      <td>0.000104</td>\n",
              "      <td>23.960745</td>\n",
              "      <td>21.590840</td>\n",
              "      <td>21.264570</td>\n",
              "      <td>21.188495</td>\n",
              "      <td>23.387875</td>\n",
              "      <td>21.943090</td>\n",
              "      <td>21.662500</td>\n",
              "      <td>21.667430</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3651928</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>19.75852</td>\n",
              "      <td>19.35878</td>\n",
              "      <td>19.18816</td>\n",
              "      <td>19.27560</td>\n",
              "      <td>19.14605</td>\n",
              "      <td>20.10190</td>\n",
              "      <td>19.68995</td>\n",
              "      <td>19.53116</td>\n",
              "      <td>19.63202</td>\n",
              "      <td>19.48096</td>\n",
              "      <td>0.617476</td>\n",
              "      <td>1.415241</td>\n",
              "      <td>0.599036</td>\n",
              "      <td>1.363397</td>\n",
              "      <td>19.18816</td>\n",
              "      <td>19.27560</td>\n",
              "      <td>19.14605</td>\n",
              "      <td>0.906366</td>\n",
              "      <td>0.000071</td>\n",
              "      <td>19.558650</td>\n",
              "      <td>19.273470</td>\n",
              "      <td>19.231880</td>\n",
              "      <td>19.210825</td>\n",
              "      <td>19.895925</td>\n",
              "      <td>19.610555</td>\n",
              "      <td>19.581590</td>\n",
              "      <td>19.556490</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2920238 rows × 28 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                objid  modelMag_u  modelMag_g  ...       mfgr       mfri       mfiz\n",
              "2        1.237680e+18    23.74654    23.19651  ...  23.060810  21.895220  20.826330\n",
              "8        1.237680e+18    22.18754    20.26853  ...  20.431030  19.300000  18.775600\n",
              "12       1.237680e+18    19.85871    19.78292  ...  19.977540  19.884490  19.771875\n",
              "15       1.237680e+18    20.89670    20.47772  ...  20.712625  20.602395  20.359570\n",
              "16       1.237680e+18    20.44421    20.43502  ...  20.704615  20.656050  20.329845\n",
              "...               ...         ...         ...  ...        ...        ...        ...\n",
              "3651918  1.237680e+18    23.55071    22.16176  ...  22.409755  21.511060  20.572055\n",
              "3651919  1.237680e+18    21.86682    21.68359  ...  21.877690  21.498485  21.041875\n",
              "3651923  1.237680e+18    22.31663    21.52032  ...  21.718615  21.318625  21.207190\n",
              "3651924  1.237680e+18    26.08879    21.83270  ...  21.943090  21.662500  21.667430\n",
              "3651928  1.237680e+18    19.75852    19.35878  ...  19.610555  19.581590  19.556490\n",
              "\n",
              "[2920238 rows x 28 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        },
        "id": "FHhhdDHNaZGZ",
        "outputId": "1d3bf94f-bae3-47e9-d9bc-b32e9b9fc89e"
      },
      "source": [
        "dataset.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>objid</th>\n",
              "      <th>modelMag_u</th>\n",
              "      <th>modelMag_g</th>\n",
              "      <th>modelMag_r</th>\n",
              "      <th>modelMag_i</th>\n",
              "      <th>modelMag_z</th>\n",
              "      <th>fiberMag_u</th>\n",
              "      <th>fiberMag_g</th>\n",
              "      <th>fiberMag_r</th>\n",
              "      <th>fiberMag_i</th>\n",
              "      <th>fiberMag_z</th>\n",
              "      <th>petroR50_r</th>\n",
              "      <th>petroR90_r</th>\n",
              "      <th>petroR50_z</th>\n",
              "      <th>petroR90_z</th>\n",
              "      <th>r</th>\n",
              "      <th>i</th>\n",
              "      <th>z</th>\n",
              "      <th>redshift</th>\n",
              "      <th>zerr</th>\n",
              "      <th>mmug</th>\n",
              "      <th>mmgr</th>\n",
              "      <th>mmri</th>\n",
              "      <th>mmiz</th>\n",
              "      <th>mfug</th>\n",
              "      <th>mfgr</th>\n",
              "      <th>mfri</th>\n",
              "      <th>mfiz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>21.63269</td>\n",
              "      <td>21.27911</td>\n",
              "      <td>21.22550</td>\n",
              "      <td>20.95690</td>\n",
              "      <td>20.82752</td>\n",
              "      <td>22.05036</td>\n",
              "      <td>21.62439</td>\n",
              "      <td>21.64663</td>\n",
              "      <td>21.32027</td>\n",
              "      <td>21.40336</td>\n",
              "      <td>0.562805</td>\n",
              "      <td>1.068754</td>\n",
              "      <td>0.166565</td>\n",
              "      <td>0.238479</td>\n",
              "      <td>21.22550</td>\n",
              "      <td>20.95690</td>\n",
              "      <td>20.82752</td>\n",
              "      <td>1.839604</td>\n",
              "      <td>0.001146</td>\n",
              "      <td>21.455900</td>\n",
              "      <td>21.252305</td>\n",
              "      <td>21.091200</td>\n",
              "      <td>20.89221</td>\n",
              "      <td>21.837375</td>\n",
              "      <td>21.635510</td>\n",
              "      <td>21.48345</td>\n",
              "      <td>21.361815</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>19.74829</td>\n",
              "      <td>19.45819</td>\n",
              "      <td>19.33118</td>\n",
              "      <td>19.01371</td>\n",
              "      <td>18.80369</td>\n",
              "      <td>20.10190</td>\n",
              "      <td>19.79539</td>\n",
              "      <td>19.66580</td>\n",
              "      <td>19.35386</td>\n",
              "      <td>19.15471</td>\n",
              "      <td>0.631004</td>\n",
              "      <td>1.339584</td>\n",
              "      <td>0.602479</td>\n",
              "      <td>1.350725</td>\n",
              "      <td>19.33118</td>\n",
              "      <td>19.01371</td>\n",
              "      <td>18.80369</td>\n",
              "      <td>2.069321</td>\n",
              "      <td>0.000354</td>\n",
              "      <td>19.603240</td>\n",
              "      <td>19.394685</td>\n",
              "      <td>19.172445</td>\n",
              "      <td>18.90870</td>\n",
              "      <td>19.948645</td>\n",
              "      <td>19.730595</td>\n",
              "      <td>19.50983</td>\n",
              "      <td>19.254285</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>23.74654</td>\n",
              "      <td>23.19651</td>\n",
              "      <td>22.01303</td>\n",
              "      <td>20.80744</td>\n",
              "      <td>19.91660</td>\n",
              "      <td>24.60136</td>\n",
              "      <td>23.64122</td>\n",
              "      <td>22.48040</td>\n",
              "      <td>21.31004</td>\n",
              "      <td>20.34262</td>\n",
              "      <td>0.645596</td>\n",
              "      <td>1.295767</td>\n",
              "      <td>0.923503</td>\n",
              "      <td>2.043880</td>\n",
              "      <td>22.01303</td>\n",
              "      <td>20.80744</td>\n",
              "      <td>19.91660</td>\n",
              "      <td>0.814588</td>\n",
              "      <td>0.000212</td>\n",
              "      <td>23.471525</td>\n",
              "      <td>22.604770</td>\n",
              "      <td>21.410235</td>\n",
              "      <td>20.36202</td>\n",
              "      <td>24.121290</td>\n",
              "      <td>23.060810</td>\n",
              "      <td>21.89522</td>\n",
              "      <td>20.826330</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>20.63075</td>\n",
              "      <td>20.25426</td>\n",
              "      <td>20.04478</td>\n",
              "      <td>19.79378</td>\n",
              "      <td>19.70326</td>\n",
              "      <td>20.99431</td>\n",
              "      <td>20.60174</td>\n",
              "      <td>20.40031</td>\n",
              "      <td>20.19795</td>\n",
              "      <td>19.95800</td>\n",
              "      <td>0.657796</td>\n",
              "      <td>1.482055</td>\n",
              "      <td>0.764345</td>\n",
              "      <td>1.480650</td>\n",
              "      <td>20.04478</td>\n",
              "      <td>19.79378</td>\n",
              "      <td>19.70326</td>\n",
              "      <td>1.455912</td>\n",
              "      <td>0.000394</td>\n",
              "      <td>20.442505</td>\n",
              "      <td>20.149520</td>\n",
              "      <td>19.919280</td>\n",
              "      <td>19.74852</td>\n",
              "      <td>20.798025</td>\n",
              "      <td>20.501025</td>\n",
              "      <td>20.29913</td>\n",
              "      <td>20.077975</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.237680e+18</td>\n",
              "      <td>21.38126</td>\n",
              "      <td>21.13488</td>\n",
              "      <td>21.09993</td>\n",
              "      <td>20.92882</td>\n",
              "      <td>20.80634</td>\n",
              "      <td>21.85225</td>\n",
              "      <td>21.45860</td>\n",
              "      <td>21.45340</td>\n",
              "      <td>21.19782</td>\n",
              "      <td>20.63075</td>\n",
              "      <td>0.728897</td>\n",
              "      <td>2.212801</td>\n",
              "      <td>1.027675</td>\n",
              "      <td>1.458456</td>\n",
              "      <td>21.09993</td>\n",
              "      <td>20.92882</td>\n",
              "      <td>20.80634</td>\n",
              "      <td>1.651130</td>\n",
              "      <td>0.000760</td>\n",
              "      <td>21.258070</td>\n",
              "      <td>21.117405</td>\n",
              "      <td>21.014375</td>\n",
              "      <td>20.86758</td>\n",
              "      <td>21.655425</td>\n",
              "      <td>21.456000</td>\n",
              "      <td>21.32561</td>\n",
              "      <td>20.914285</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          objid  modelMag_u  modelMag_g  ...       mfgr      mfri       mfiz\n",
              "0  1.237680e+18    21.63269    21.27911  ...  21.635510  21.48345  21.361815\n",
              "1  1.237680e+18    19.74829    19.45819  ...  19.730595  19.50983  19.254285\n",
              "2  1.237680e+18    23.74654    23.19651  ...  23.060810  21.89522  20.826330\n",
              "3  1.237680e+18    20.63075    20.25426  ...  20.501025  20.29913  20.077975\n",
              "4  1.237680e+18    21.38126    21.13488  ...  21.456000  21.32561  20.914285\n",
              "\n",
              "[5 rows x 28 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6CfX4iuaaaS9"
      },
      "source": [
        "#these columns are not required\n",
        "dataset.drop(['objid'], axis = 1,inplace=True) #dropping objid column\n",
        "dataset.drop(['zerr'], axis = 1,inplace=True) #dropping zerr column"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9ZJ8jN1jamBl"
      },
      "source": [
        "dataset[dataset[\"modelMag_u\"]==-9999]=np.nan\n",
        "dataset[dataset[\"modelMag_g\"]==-9999]=np.nan\n",
        "dataset[dataset[\"modelMag_r\"]==-9999]=np.nan\n",
        "dataset[dataset[\"modelMag_i\"]==-9999]=np.nan\n",
        "dataset[dataset[\"modelMag_z\"]==-9999]=np.nan\n",
        "dataset[dataset[\"fiberMag_u\"]==-9999]=np.nan\n",
        "dataset[dataset[\"fiberMag_g\"]==-9999]=np.nan\n",
        "dataset[dataset[\"fiberMag_r\"]==-9999]=np.nan\n",
        "dataset[dataset[\"fiberMag_i\"]==-9999]=np.nan\n",
        "dataset[dataset[\"fiberMag_z\"]==-9999]=np.nan\n",
        "dataset[dataset[\"petroR50_r\"]==-9999]=np.nan\n",
        "dataset[dataset[\"petroR90_r\"]==-9999]=np.nan\n",
        "dataset[dataset[\"petroR50_z\"]==-9999]=np.nan\n",
        "dataset[dataset[\"petroR90_z\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mmug\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mmgr\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mmri\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mmiz\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mfug\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mfgr\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mfri\"]==-9999]=np.nan\n",
        "dataset[dataset[\"mfiz\"]==-9999]=np.nan\n",
        "dataset[dataset[\"r\"]==-9999]=np.nan\n",
        "dataset[dataset[\"i\"]==-9999]=np.nan\n",
        "dataset[dataset[\"z\"]==-9999]=np.nan\n",
        "dataset[dataset[\"redshift\"]==-9999]=np.nan\n",
        "dataset[dataset[\"redshift\"]<0]=np.nan"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "egxJXpvEapF3",
        "outputId": "4c961e53-f6fd-445f-8bf4-f30977fad522"
      },
      "source": [
        "#Dropping null values\n",
        "dataset=dataset.dropna()\n",
        "dataset.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3548153, 26)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "9srDDSAmar_Z",
        "outputId": "089374cf-cdc2-4f41-db70-dadfcf4adceb"
      },
      "source": [
        "#checking the data distribution\n",
        "plt.hist(dataset['redshift'], bins=50, alpha=0.7)\n",
        "plt.xlabel('SpecZ')\n",
        "plt.ylabel('Number of samples')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Number of samples')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEJCAYAAABYCmo+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAcj0lEQVR4nO3df7RdZX3n8feHX8ogmCDXrAzBBscsLMUB4QJxcPBHFiGgQ1hWEVaVwGSIDmhxdE0FR00LtcV2VQotpM2QSNJRELFK6gRDFojWNQVyAwzIrxIpDIlAAgGCMuCAn/ljPxcPl3PP3Un2OSf35vNaa6+z93f/eJ7rcvHN82M/W7aJiIho0i79rkBEREw8SS4REdG4JJeIiGhckktERDQuySUiIhqX5BIREY3rWnKRdJCkO1u2LZI+I2lfSaslPVh+J5frJelSSesk3SXp8JZnzSvXPyhpXkv8CEl3l3sulaQSb1tGRET0RteSi+0HbB9m+zDgCOB54LvAecCNtmcAN5ZjgBOAGWVbACyCKlEAC4GjgaOAhS3JYhFwVst9c0p8tDIiIqIHdutRObOAn9l+RNJc4L0lvgy4Gfg8MBdY7uqtzlskTZI0tVy72vZmAEmrgTmSbgb2sX1LiS8HTgauL89qV8ao9ttvP0+fPn37/9KIiJ3I2rVrn7Q9MDLeq+RyKnBV2Z9i+7Gy/zgwpezvDzzacs/6EusUX98m3qmMUU2fPp2hoaFaf0xERFQkPdIu3vUBfUl7ACcB3x55rrRSurr+TKcyJC2QNCRpaNOmTd2sRkTETqUXs8VOAG63/UQ5fqJ0d1F+N5b4BuCAlvumlVin+LQ28U5lvIrtxbYHbQ8ODLymVRcREduoF8nlNH7TJQawAhie8TUPuK4lfnqZNTYTeLZ0ba0CZkuaXAbyZwOryrktkmaWWWKnj3hWuzIiIqIHujrmImkv4DjgEy3hi4BrJM0HHgFOKfGVwInAOqqZZWcC2N4s6UJgTbnuguHBfeBs4EpgT6qB/OvHKCMiInpAWXK/Mjg46AzoR0RsHUlrbQ+OjOcN/YiIaFySS0RENC7JJSIiGpfkEhERjevVG/pRw/wr17SNLznjyB7XJCJi+6TlEhERjUtyiYiIxiW5RERE45JcIiKicUkuERHRuCSXiIhoXJJLREQ0LsklIiIal+QSERGNS3KJiIjGJblERETjklwiIqJxSS4REdG4JJeIiGhcltxvQJbKj4h4tbRcIiKicUkuERHRuK4mF0mTJF0r6X5J90l6l6R9Ja2W9GD5nVyulaRLJa2TdJekw1ueM69c/6CkeS3xIyTdXe65VJJKvG0ZERHRG91uuVwC/MD224FDgfuA84Abbc8AbizHACcAM8q2AFgEVaIAFgJHA0cBC1uSxSLgrJb75pT4aGVEREQPdC25SHojcCywBMD2r2w/A8wFlpXLlgEnl/25wHJXbgEmSZoKHA+str3Z9tPAamBOObeP7VtsG1g+4lntyoiIiB7oZsvlQGAT8HVJd0i6QtJewBTbj5VrHgemlP39gUdb7l9fYp3i69vE6VBGRET0QDeTy27A4cAi2+8EfsmI7qnS4nAX69CxDEkLJA1JGtq0aVM3qxERsVPpZnJZD6y3fWs5vpYq2TxRurQovxvL+Q3AAS33TyuxTvFpbeJ0KONVbC+2PWh7cGBgYJv+yIiIeK2uJRfbjwOPSjqohGYB9wIrgOEZX/OA68r+CuD0MmtsJvBs6dpaBcyWNLkM5M8GVpVzWyTNLLPETh/xrHZlRERED3T7Df1PA9+QtAfwEHAmVUK7RtJ84BHglHLtSuBEYB3wfLkW25slXQgMvwZ/ge3NZf9s4EpgT+D6sgFcNEoZERHRA11NLrbvBAbbnJrV5loD54zynKXA0jbxIeCQNvGn2pURERG9kTf0IyKicUkuERHRuCSXiIhoXJJLREQ0LsklIiIal+QSERGNS3KJiIjGJblERETjklwiIqJxSS4REdG4JJeIiGhckktERDQuySUiIhqX5BIREY1LcomIiMYluUREROOSXCIionFJLhER0bgkl4iIaFySS0RENC7JJSIiGpfkEhERjetqcpH0sKS7Jd0paajE9pW0WtKD5XdyiUvSpZLWSbpL0uEtz5lXrn9Q0ryW+BHl+evKvepURkRE9EYvWi7vs32Y7cFyfB5wo+0ZwI3lGOAEYEbZFgCLoEoUwELgaOAoYGFLslgEnNVy35wxyoiIiB4YM7lIOlfSPqVlsUTS7ZJmb0eZc4FlZX8ZcHJLfLkrtwCTJE0FjgdW295s+2lgNTCnnNvH9i22DSwf8ax2ZURERA/Uabn8R9tbgNnAZODjwEU1n2/gBklrJS0osSm2Hyv7jwNTyv7+wKMt964vsU7x9W3incqIiIge2K3GNSq/JwJ/Z/ue4bGNGt5te4OkNwOrJd3fetK2JXkr6rvVOpVREt4CgLe85S3drEZExE6lTstlraQbqJLLKkl7A7+u83DbG8rvRuC7VGMmT5QuLcrvxnL5BuCAltunlVin+LQ2cTqUMbJ+i20P2h4cGBio8ydFREQNdZLLfKoB8SNtPw/sAZw51k2S9iqJCEl7UXWr/RRYAQzP+JoHXFf2VwCnl7GdmcCzpWtrFTBb0uQykD8bWFXObZE0s7SkTh/xrHZlRERED9TpFjNwMPBB4AJgL+D1Ne6bAny39KDtBnzT9g8krQGukTQfeAQ4pVy/kqp1tA54npLAbG+WdCGwplx3ge3NZf9s4EpgT+D6skE1JtSujIiI6IE6yeVyqm6w91Mll+eA7wBHdrrJ9kPAoW3iTwGz2sQNnDPKs5YCS9vEh4BD6pYRERG9USe5HG37cEl3ANh+WtIeXa5XtJh/5Zq28SVndMzvERF9U2fM5f9J2pWqewxJA9Qc0I+IiJ1TneRyKdVMrzdL+grwE+BPulqriIgY18bsFrP9DUlrqcYwBJxs+76u1ywiIsatUZNLWdNr2EbgqtZzLTO2IiIiXqVTy2Ut1ThLu7fxDby1KzWKiIhxb9TkYvvAXlYkIiImjjpTkZH0IeDdVC2Wf7T9va7WKiIixrU6S+5fDnwSuJtq+ZZPSrqs2xWLiIjxq07L5f3Ab5c36JG0DLinq7WKiIhxrc57LuuA1vXoDyixiIiItuq0XPYG7pN0Wzk+EhiStALA9kndqlxERIxPdZLLl7tei4iImFDqvKH/IwBJ+7Ren5coIyJiNGMml/Ip4AuAF6gWrBR5iTIiIjqo0y32X4FDbD/Z7cpERMTEUGe22M+ovgwZERFRS52Wy/nA/5J0K/DicND273etVhERMa7VSS5/C9xE9YZ+PhIWERFjqpNcdrf92a7XJCIiJow6Yy7XS1ogaaqkfYe3rtcsIiLGrTotl9PK7/ktsUxFjoiIUY3ZcrF9YJutdmKRtKukOyR9vxwfKOlWSeskfUvSHiX+unK8rpyf3vKM80v8AUnHt8TnlNg6See1xNuWERERvVGnWwxJh0g6RdLpw9tWlHEucF/L8VeBi22/DXgamF/i84GnS/zich2SDgZOBX4HmANcXhLWrsBlwAnAwcBp5dpOZURERA/U+Z7LQuCvyvY+4M+AWotVSpoGfAC4ohyLagn/a8sly4CTy/7cckw5P6tcPxe42vaLtv+FakXmo8q2zvZDtn8FXA3MHaOMiIjogTotlw8Ds4DHbZ8JHAq8sebz/xL4A34zhflNwDO2XyrH64H9y/7+wKMA5fyz5fpX4iPuGS3eqYxXKRMVhiQNbdq0qeafFBERY6mTXP6v7V8DL5XFKzdSfdOlI0kfBDbaXruddewa24ttD9oeHBgY6Hd1IiImjDqzxYYkTQL+O7AW+AXwTzXuOwY4SdKJwOuBfYBLgEmSdisti2nAhnL9BqqktV7SblSto6da4sNa72kXf6pDGRER0QN1ZoudbfsZ238DHAfMK91jY913vu1ptqdTDcjfZPv3gB9SdbUBzAOuK/sryjHl/E3l08orgFPLbLIDgRnAbcAaYEaZGbZHKWNFuWe0MiIiogfqDOgfI2mvcvhu4AxJv7UdZX4e+KykdVTjI0tKfAnwphL/LHAegO17gGuAe4EfAOfYfrm0Sj4FrKKajXZNubZTGRER0QN1usUWAYdKOhT4HNXMr+XAe+oWYvtm4Oay/xDVTK+R17wAfGSU+78CfKVNfCWwsk28bRkREdEbdQb0XypdTXOBv7Z9GbB3d6sVERHjWZ2Wy3OSzgc+BhwraRdg9+5WKyIixrM6LZePUn3HZb7tx6lmX/15V2sVERHj2pgtl5JQvtZy/H+oxlwiIiLaqrW2WERExNZIcomIiMaNmlwk3Vh+v9q76kRExETQacxlqqR/R7WEy9WAWk/avr2rNYuIiHGrU3L5MvAlqtlhXxtxzlTL2kdERLzGqMnF9rXAtZK+ZPvCHtYpIiLGuTpTkS+UdBJwbAndbPv73a3WxDD/yjX9rkJERF/UWbjyT6k+VXxv2c6V9CfdrlhERIxfdZZ/+QBwWPlgGJKWAXcAX+hmxSIiYvyq+57LpJb9up84joiInVSdlsufAndI+iHVdORjKd9aiYiIaKfOgP5Vkm4Gjiyhz5f1xiIiItqq03LB9mNUnxuOiIgYU9YWi4iIxiW5RERE4zomF0m7Srq/V5WJiIiJoWNysf0y8ICkt/SoPhERMQHUGdCfDNwj6Tbgl8NB2yd1rVYRETGu1UkuX9qWB0t6PfBj4HWlnGttL5R0IHA18CZgLfBx27+S9DqqzycfATwFfNT2w+VZ5wPzgZeB37e9qsTnAJcAuwJX2L6oxNuWsS1/R0REbL0xB/Rt/wh4GNi97K8B6nzL5UXg/bYPBQ4D5kiaCXwVuNj224CnqZIG5ffpEr+4XIekg4FTgd8B5gCXl7GgXYHLgBOAg4HTyrV0KCMiInqgzsKVZwHXAn9bQvsD3xvrPld+UQ53L9vwd2CuLfFlwMllf245ppyfJUklfrXtF23/C7AOOKps62w/VFolVwNzyz2jlRERET1QZyryOcAxwBYA2w8Cb67z8NLCuBPYCKwGfgY8Y/ulcsl6qmRF+X20lPES8CxVt9Yr8RH3jBZ/U4cyRtZvgaQhSUObNm2q8ydFREQNdZLLi63jFZJ2o2qBjMn2y7YPo/qa5VHA27epll1ie7HtQduDAwMD/a5ORMSEUSe5/EjSF4A9JR0HfBv4h60pxPYzwA+BdwGTSoKCKulsKPsbgAPglQT2RqqB/VfiI+4ZLf5UhzIiIqIH6iSX84BNwN3AJ4CVwBfHuknSgKRJZX9P4DjgPqok8+Fy2TzgurK/ohxTzt9k2yV+qqTXlVlgM4DbqCYWzJB0oKQ9qAb9V5R7RisjIiJ6oM6qyL8uHwi7lao77IHyH/CxTAWWlVlduwDX2P6+pHuBqyX9MdVHx5aU65cAfydpHbCZKllg+x5J11B9BfMl4JzycieSPgWsopqKvNT2PeVZnx+ljIiI6IExk4ukDwB/QzUYL+BASZ+wfX2n+2zfBbyzTfwhqvGXkfEXgI+M8qyvAF9pE19J1ZKqVUZERPRGnZco/wJ4n+11AJL+DfA/gY7JJSIidl51xlyeG04sxUPAc12qT0RETACjtlwkfajsDklaCVxDNebyEarB9IiIiLY6dYv9h5b9J4D3lP1NwJ5dq1FERIx7oyYX22f2siIRETFx1JktdiDwaWB66/VZcj8iIkZTZ7bY96jeE/kH4NfdrU5EREwEdZLLC7Yv7XpNIiJiwqiTXC6RtBC4geobLQDYrvNNl4iI2AnVSS7vAD5O9Y2U4W6x4e+yREREvEad5PIR4K35THBERNRVJ7n8FJhE9cGv2IHMv3L0d1mXnHFkD2sSEfFqdZLLJOB+SWt49ZhLpiJHRERbdZLLwq7XIiIiJpQ633P5US8qEhERE0edN/Sfo5odBrAHsDvwS9v7dLNiERExftVpuew9vC9JwFxgZjcrFRER41ud77m8wpXvAcd3qT4RETEB1OkW+1DL4S7AIPBC12oUERHjXp3ZYq3fdXkJeJiqaywiIqKtOmMu+a5LRERslU6fOf5yh/ts+8JOD5Z0ALAcmEI122yx7Usk7Qt8i+r7MA8Dp9h+ukwWuAQ4EXgeOGN4cUxJ84Avlkf/se1lJX4EcCXVlzFXAufa9mhldKpvREQ0p9OA/i/bbADzgc/XePZLwOdsH0w1u+wcSQcD5wE32p4B3FiOAU4AZpRtAbAIoCSKhcDRwFHAQkmTyz2LgLNa7ptT4qOVERERPTBqcrH9F8MbsJiqdXAmcDXw1rEebPux4ZaH7eeA+4D9qcZrlpXLlgEnl/25wPIyI+0WYJKkqVQz01bb3lxaH6uBOeXcPrZvsW2qVlLrs9qVERERPdBxzKW0Gj4L/B7Vf6QP35buJUnTgXcCtwJTbD9WTj1O1W0GVeJ5tOW29SXWKb6+TZwOZURERA+M2nKR9OfAGuA54B22/3AbE8sbgO8An7G9pfVcaXG47Y0N6VSGpAWShiQNbdq0qZvViIjYqXQac/kc8K+pBtJ/LmlL2Z6TtKXDfa+QtDtVYvmG7b8v4SdKlxbld3gp/w3AAS23TyuxTvFpbeKdyngV24ttD9oeHBgYqPMnRUREDZ3GXHaxvaftvW3v07LtXWddsTL7awlwn+2vtZxaAcwr+/OA61rip6syE3i2dG2tAmZLmlwG8mcDq8q5LZJmlrJOH/GsdmVEREQP1HmJclsdQ/V55Lsl3VliXwAuAq6RNB94BDilnFtJNQ15HdVU5DMBbG+WdCFVFx3ABbY3l/2z+c1U5OvLRocyIiKiB7qWXGz/BNAop2e1ud7AOaM8aymwtE18CDikTfypdmVERERvbNXClREREXUkuUREROOSXCIionFJLhER0bgkl4iIaFySS0RENC7JJSIiGpfkEhERjUtyiYiIxiW5RERE45JcIiKicUkuERHRuCSXiIhoXJJLREQ0LsklIiIal+QSERGNS3KJiIjGdfMzx9FH869c0za+5Iwje1yTiNgZpeUSERGNS3KJiIjGJblERETjklwiIqJxXUsukpZK2ijppy2xfSWtlvRg+Z1c4pJ0qaR1ku6SdHjLPfPK9Q9KmtcSP0LS3eWeSyWpUxkREdE73Wy5XAnMGRE7D7jR9gzgxnIMcAIwo2wLgEVQJQpgIXA0cBSwsCVZLALOarlvzhhlREREj3Qtudj+MbB5RHgusKzsLwNObokvd+UWYJKkqcDxwGrbm20/DawG5pRz+9i+xbaB5SOe1a6MiIjokV6PuUyx/VjZfxyYUvb3Bx5tuW59iXWKr28T71RGRET0SN8G9EuLw/0sQ9ICSUOShjZt2tTNqkRE7FR6nVyeKF1alN+NJb4BOKDlumkl1ik+rU28UxmvYXux7UHbgwMDA9v8R0VExKv1OrmsAIZnfM0DrmuJn15mjc0Eni1dW6uA2ZIml4H82cCqcm6LpJllltjpI57VroyIiOiRrq0tJukq4L3AfpLWU836ugi4RtJ84BHglHL5SuBEYB3wPHAmgO3Nki4EhhfKusD28CSBs6lmpO0JXF82OpQRERE90rXkYvu0UU7NanOtgXNGec5SYGmb+BBwSJv4U+3KiIiI3skb+hER0bgkl4iIaFySS0RENC4fC9vJ5CNiEdELablERETjklwiIqJx6RYLIN1lEdGsJJfoKEknIrZFkktsk9GSzmiSjCJ2LhlziYiIxqXlEj2R7rWInUuSS/RVkk7ExJTkEuPK1o71QBJVRD8kucQOaVuSSETsODKgHxERjUvLJSa8jOtE9F5aLhER0bgkl4iIaFy6xWKnle6yiO5JyyUiIhqXlkvECGnRRGy/JJeImpJ0IuqbsMlF0hzgEmBX4ArbF/W5SjFBJelEvNaETC6SdgUuA44D1gNrJK2wfW9/axY7kySd2JlNyOQCHAWss/0QgKSrgblAkkv0XZJO7AwmanLZH3i05Xg9cHSf6hJRSz/XU0ti23E19Y+RXv+jRra78uB+kvRhYI7t/1SOPw4cbftTI65bACwohwcBD2xjkfsBT27jvb2WunbPeKpv6todO2Ndf8v2wMjgRG25bAAOaDmeVmKvYnsxsHh7C5M0ZHtwe5/TC6lr94yn+qau3ZG6/sZEfYlyDTBD0oGS9gBOBVb0uU4RETuNCdlysf2SpE8Bq6imIi+1fU+fqxURsdOYkMkFwPZKYGWPitvurrUeSl27ZzzVN3XtjtS1mJAD+hER0V8TdcwlIiL6KMllO0maI+kBSeskndfv+oxG0lJJGyX9tN91GYukAyT9UNK9ku6RdG6/6zQaSa+XdJuk/13q+kf9rtNYJO0q6Q5J3+93XTqR9LCkuyXdKWmo3/UZi6RJkq6VdL+k+yS9q991akfSQeV/0+Fti6TPNF5OusW2XVlm5p9pWWYGOG1HXGZG0rHAL4Dltg/pd306kTQVmGr7dkl7A2uBk3fQ/10F7GX7F5J2B34CnGv7lj5XbVSSPgsMAvvY/mC/6zMaSQ8Dg7bHxXsjkpYB/2j7ijJL9V/Zfqbf9eqk/DdsA9V7gI80+ey0XLbPK8vM2P4VMLzMzA7H9o+Bzf2uRx22H7N9e9l/DriPatWFHY4rvyiHu5dth/0Xm6RpwAeAK/pdl4lE0huBY4ElALZ/taMnlmIW8LOmEwskuWyvdsvM7JD/ERyvJE0H3gnc2t+ajK50M90JbARW295h6wr8JfAHwK/7XZEaDNwgaW1ZTWNHdiCwCfh66XK8QtJe/a5UDacCV3XjwUkuscOS9AbgO8BnbG/pd31GY/tl24dRrQRxlKQdsttR0geBjbbX9rsuNb3b9uHACcA5pWt3R7UbcDiwyPY7gV8CO+wYLEDpujsJ+HY3np/ksn1qLTMTW6+MX3wH+Ibtv+93feoo3SA/BOb0uy6jOAY4qYxlXA28X9L/6G+VRmd7Q/ndCHyXqht6R7UeWN/Sar2WKtnsyE4Abrf9RDcenuSyfbLMTBeUQfIlwH22v9bv+nQiaUDSpLK/J9Xkjvv7W6v2bJ9ve5rt6VT/X73J9sf6XK22JO1VJnNQupdmAzvsTEfbjwOPSjqohGax43/i4zS61CUGE/gN/V4YT8vMSLoKeC+wn6T1wELbS/pbq1EdA3wcuLuMZQB8oay6sKOZCiwrs252Aa6xvUNP8R0npgDfrf6dwW7AN23/oL9VGtOngW+Uf2g+BJzZ5/qMqiTs44BPdK2MTEWOiIimpVssIiIal+QSERGNS3KJiIjGJblERETjklwiIqJxSS4RPSLpv5WVk+8qq9Ee3dBzjx6xyu2dkl6Q9J+beH7EtshU5IgeKMuvfw14r+0XJe0H7GH7510o63iqNcSOsP1808+PqCMtl4jemAo8aftFANtP2v55+WbJn5Xvltwm6W3wypv/35G0pmzHlPgbJH29XH+XpN9tLaQkrcXAx5JYop+SXCJ64wbgAEn/LOlySe9pOfes7XcAf03V4gC4BLjY9pHA7/KbJfK/NHy97X8L3DSinCXA5eNoccqYoNItFtEjZYmYfw+8j2rZjfOAPwTeb/uhsljn47bfJGkj0NplNgAcBPwIONX2g22e/0ngo8As2+NhSf2YwLK2WESP2H4ZuBm4WdLdwLzhU62Xld9dgJm2X2h9Rllr6zUkvR34YrkniSX6Lt1iET1Qvls+oyV0GDD89b+Ptvz+U9m/gWohxOH7Dyu7q4FzWuKTy0KJ3wT+i+31Xah+xFZLt1hED0g6AvgrYBLwErAOWAAMAd+i+rbGi8BptteVgfnLgN+m6mH4se1Plg+oXQYcAbwM/BHwOmA5MHJF7mW2L+723xbRTpJLRB+VD3cN2n6y33WJaFK6xSIionFpuUREROPScomIiMYluUREROOSXCIionFJLhER0bgkl4iIaFySS0RENO7/AwEUCrFIfrYmAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DZ0Gyj31HhFd"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pZLdully1CMb",
        "outputId": "786b3a15-a7c1-4480-a4d7-7a94af0640bf"
      },
      "source": [
        "dataset['redshift'].value_counts(bins=10, sort=False) # counting data in a specified range based on redshift"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-0.00805, 0.705]    2474629\n",
              "(0.705, 1.41]         574322\n",
              "(1.41, 2.116]         254737\n",
              "(2.116, 2.821]        173106\n",
              "(2.821, 3.526]         49167\n",
              "(3.526, 4.231]         10443\n",
              "(4.231, 4.936]          2912\n",
              "(4.936, 5.642]          2189\n",
              "(5.642, 6.347]          2623\n",
              "(6.347, 7.052]          4025\n",
              "Name: redshift, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7S392hWIRY_g"
      },
      "source": [
        "def getClass(value):\n",
        "  if value>=0.00 and value<1.0:\n",
        "    return '1'\n",
        "  elif value>=1.00 and value<2.0:\n",
        "    return '2'\n",
        "  elif value>=2.00 and value<3.0:\n",
        "    return '3'\n",
        "  elif value>=3.00 and value<4.0:\n",
        "    return '4'\n",
        "  elif value>=4.00 and value<5.0:\n",
        "    return '5'\n",
        "  elif value>=5.00 and value<6.0:\n",
        "    return '6'\n",
        "  elif value>=6.00 and value<=7.0:\n",
        "    return '7'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MnZmbh9ab2Zr",
        "outputId": "d207d9fa-1116-430a-e2c1-786b98eb937d"
      },
      "source": [
        "##dataset['Class']=dataset['redshift'].apply(lambda x: cls(x))\n",
        "#dataset['Class']=dataset['redshift'].apply(lambda x: 0 if x<1.0 else 1)\n",
        "dataset['Class']=dataset['redshift'].apply(lambda x: getClass(x))\n",
        "dataset=dataset.dropna()\n",
        "dataset.shape\n",
        "#dataset.loc[dataset['redshift'] >0.00 , 'Class']='A'"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3547386, 27)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9blw7sBISt8c",
        "outputId": "7f647fe6-734b-4379-eeda-84193fa8408e"
      },
      "source": [
        "dataset.Class.value_counts()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    2858292\n",
              "2     411157\n",
              "3     226041\n",
              "4      39072\n",
              "5       4935\n",
              "7       4465\n",
              "6       3424\n",
              "Name: Class, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wjw5_tXEpRwy"
      },
      "source": [
        "df_1=dataset[dataset.Class=='1']\n",
        "df_2=dataset[dataset.Class=='2']\n",
        "df_3=dataset[dataset.Class=='3']\n",
        "df_4=dataset[dataset.Class=='4']\n",
        "df_5=dataset[dataset.Class=='5']\n",
        "df_6=dataset[dataset.Class=='6']\n",
        "df_7=dataset[dataset.Class=='7']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GbsIwvzrqAX3"
      },
      "source": [
        "from sklearn.utils import resample\n",
        "df_1=resample(df_1,replace=True,n_samples=513814,random_state=123) \n",
        "df_2=resample(df_2,replace=True,n_samples=513814,random_state=123) \n",
        "df_3=resample(df_3,replace=True,n_samples=513814,random_state=123) \n",
        "df_4=resample(df_4,replace=True,n_samples=513814,random_state=123) \n",
        "df_5=resample(df_5,replace=True,n_samples=513814,random_state=123) \n",
        "df_6=resample(df_6,replace=True,n_samples=513814,random_state=123) \n",
        "df_7=resample(df_7,replace=True,n_samples=513814,random_state=123) \n",
        " "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YWhYS1uIrmPh"
      },
      "source": [
        "dataset=pd.concat([df_1,df_2,df_3,df_4,df_5,df_6,df_7])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UaeW6cZrr4Ku",
        "outputId": "324ef5c0-7117-415d-dbee-e0dec949d779"
      },
      "source": [
        "dataset.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3596698, 27)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "9A7k9dkAsHuA",
        "outputId": "3250fb02-5c92-4f8e-f61a-0b685f46adea"
      },
      "source": [
        "plt.hist(dataset['redshift'], bins=50, alpha=0.7)\n",
        "plt.xlabel('SpecZ')\n",
        "plt.ylabel('Number of samples')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Number of samples')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 54
        },
        {
          "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": "NO_ZZQxM6kRV"
      },
      "source": [
        "del dataset['Class']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CqzSWTqS6tIt"
      },
      "source": [
        "X=dataset.drop('redshift',axis=1)\n",
        "Y=dataset['redshift']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SfGLFgVzO9Uw",
        "outputId": "815c7fc8-167c-4d7c-d408-f6f30b816b2b"
      },
      "source": [
        "Y.max()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "6.999911"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KcBQ0kn8PEeN",
        "outputId": "ecc29292-9a0f-4faf-fd7a-5ca6d6425bd2"
      },
      "source": [
        "Y.min()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NEI1EJeF2SAr",
        "outputId": "ef6c0b8c-59a4-49ed-b5a7-3744000a6bce"
      },
      "source": [
        "scaler = MinMaxScaler()\n",
        "X = scaler.fit_transform(X)\n",
        "print(X)\n",
        "X.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.66375676 0.61197827 0.53750891 ... 0.70085862 0.55865358 0.60254633]\n",
            " [0.71891046 0.57464808 0.51587512 ... 0.66379014 0.53720943 0.57271676]\n",
            " [0.46209821 0.44245331 0.3741607  ... 0.45471058 0.36655347 0.42050882]\n",
            " ...\n",
            " [0.56130376 0.60101241 0.55881407 ... 0.7139768  0.63467415 0.69779779]\n",
            " [0.58673038 0.67529803 0.59342615 ... 0.7724896  0.62715715 0.65735433]\n",
            " [0.52469998 0.57536355 0.5584971  ... 0.65070576 0.59626463 0.66339623]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3596698, 25)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W695P9kt2ThV",
        "outputId": "33605889-a809-44dc-ce1c-608c8061e3d6"
      },
      "source": [
        "# Target vector\n",
        "y = dataset['redshift']\n",
        "print(y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "982048     0.498486\n",
            "356674     0.560501\n",
            "36822      0.148298\n",
            "1584991    0.074713\n",
            "2492476    0.681300\n",
            "             ...   \n",
            "3635284    6.173983\n",
            "1983692    6.549440\n",
            "3512759    6.695148\n",
            "564131     6.820958\n",
            "2920048    6.139973\n",
            "Name: redshift, Length: 3596698, dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5cM_U-Q_Ff8c",
        "outputId": "bfd13fd7-f5dd-486c-80d2-35b6df2a92d3"
      },
      "source": [
        "y"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "982048     0.498486\n",
              "356674     0.560501\n",
              "36822      0.148298\n",
              "1584991    0.074713\n",
              "2492476    0.681300\n",
              "             ...   \n",
              "3635284    6.173983\n",
              "1983692    6.549440\n",
              "3512759    6.695148\n",
              "564131     6.820958\n",
              "2920048    6.139973\n",
              "Name: redshift, Length: 3596698, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QOniKt7I2XRI",
        "outputId": "f5a0a8d1-a668-428e-a6ab-1e4f31877c6d"
      },
      "source": [
        "y=y.to_numpy() #pd needs to convert to numpy before applying minmaxscaler\n",
        "y=y.reshape(-1,1) # this is for using MinMAxscalerr\n",
        "print(y.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(3596698, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EcwE6tft2acv",
        "outputId": "ec50c3bc-6e18-4865-8d0e-f3ff5d938201"
      },
      "source": [
        "scaler = MinMaxScaler()\n",
        "y = scaler.fit_transform(y)\n",
        "print(y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.07121319]\n",
            " [0.08007258]\n",
            " [0.02118568]\n",
            " ...\n",
            " [0.95646188]\n",
            " [0.97443496]\n",
            " [0.87715015]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6tGUOMac2dxH"
      },
      "source": [
        "#user dfined function for train,test and validation\n",
        "def my_train_test_split(X, y, ratio_train, ratio_val, seed=42):\n",
        "    idx = np.arange(X.shape[0])\n",
        "    np.random.seed(seed)\n",
        "    np.random.shuffle(idx)\n",
        "\n",
        "    limit_train = int(ratio_train * X.shape[0])\n",
        "    limit_val = int((ratio_train + ratio_val) * X.shape[0])\n",
        "\n",
        "    idx_train = idx[:limit_train]\n",
        "    idx_val = idx[limit_train:limit_val]\n",
        "    idx_test = idx[limit_val:]\n",
        "\n",
        "    X_train, y_train = X[idx_train], y[idx_train]\n",
        "    X_val, y_val = X[idx_val], y[idx_val]\n",
        "    X_test, y_test = X[idx_test], y[idx_test]\n",
        "\n",
        "    return X_train, X_val, X_test, y_train, y_val, y_test"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q-Wn4NoX2iXN"
      },
      "source": [
        "X_train, X_val, X_test, y_train, y_val, y_test = my_train_test_split(X, y, .75,.15,seed=42)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DbXrMBnB2oDM",
        "outputId": "e8048af6-3da8-44a5-8a2a-e5ac0f693b86"
      },
      "source": [
        "print(X_train.shape)\n",
        "print(X_val.shape)\n",
        "print(X_test.shape)\n",
        "print(y_train.shape)\n",
        "print(y_val.shape)\n",
        "print(y_test.shape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(2697523, 25)\n",
            "(539505, 25)\n",
            "(359670, 25)\n",
            "(2697523, 1)\n",
            "(539505, 1)\n",
            "(359670, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "F_XCGm4TeBCY"
      },
      "source": [
        "y_train=y_train.ravel()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BEu6rRXEeb4p",
        "outputId": "59d48c17-a18d-469c-90fd-108722409371"
      },
      "source": [
        "y_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2697523,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "58ThmLufV5A3",
        "outputId": "fbc7d408-dd45-4e55-da71-282bf0cb9e97"
      },
      "source": [
        "#code for xtratree regressor\n",
        "from sklearn.datasets import make_regression\n",
        "from sklearn.ensemble import ExtraTreesRegressor\n",
        "import timeit\n",
        "start = timeit.default_timer()\n",
        "model = ExtraTreesRegressor(n_estimators=50, random_state=0)\n",
        "model.fit(X_train, y_train)\n",
        "stop = timeit.default_timer()\n",
        "print('Time: ', (stop - start)/60,'minutes')  \n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Time:  16.827047789483334 minutes\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hu06O-6QmT18",
        "outputId": "d954b8d6-b94c-4447-b789-ee1eff45d2e7"
      },
      "source": [
        "y_pred_tree = model.predict(X_test)\n",
        "from sklearn import metrics\n",
        "from math import sqrt\n",
        "from sklearn.metrics import mean_squared_error\n",
        "var = metrics.explained_variance_score(y_test, y_pred_tree)\n",
        "mae = metrics.mean_absolute_error(y_test, y_pred_tree)\n",
        "mse = metrics.mean_squared_error(y_test, y_pred_tree)\n",
        "r2 = metrics.r2_score(y_test, y_pred_tree)\n",
        "rmse = sqrt(mean_squared_error(y_test, y_pred_tree))\n",
        "print ('Explained variance (best possible score is 1.0, lower values are worse):', var)\n",
        "print ('Mean Absolute Error (best is 0.0):', mae)\n",
        "print ('Mean Squared Error (best is 0.0):', mse)\n",
        "print ('R2 score (best is 1.0):', r2)\n",
        "print ('root Mean Squared Error (best is 0.0):', rmse)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Explained variance (best possible score is 1.0, lower values are worse): 0.9923523713518858\n",
            "Mean Absolute Error (best is 0.0): 0.006432421829877173\n",
            "Mean Squared Error (best is 0.0): 0.0006700882832036021\n",
            "R2 score (best is 1.0): 0.9921968742036995\n",
            "root Mean Squared Error (best is 0.0): 0.025886063493772128\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PvjaIZa0AgC4",
        "outputId": "679d7c46-8354-437d-9f22-2152ee8efaa7"
      },
      "source": [
        "y_pred_tree"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.93430705, 0.99516788, 0.59575037, ..., 0.69537384, 0.85295113,\n",
              "       0.58018009])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KrP_Gq2Ighxe",
        "outputId": "cd56b362-63ba-4858-f400-a6b40f07ac71"
      },
      "source": [
        "y_test_tree=y_test\n",
        "y_test_tree=pd.DataFrame(y_test_tree)\n",
        "y_pred_tree=pd.DataFrame(y_pred_tree)\n",
        "outl=y_pred_tree[abs(y_test_tree-y_pred_tree)>0.1]\n",
        "outl=outl.dropna()\n",
        "print('Catastropic Outliers: ', outl.shape[0]*100.0/y_test.shape[0],'%')\n",
        "print('Total rms: ', np.sqrt(np.mean((y_test_tree-y_pred_tree)**2)))\n",
        "print('rms w/o outliers', np.sqrt(np.mean(((y_test_tree-y_pred_tree)[abs(y_test_tree-y_pred_tree)<0.1])**2)))\n",
        "print('Bias: ', np.mean(y_test_tree-y_pred_tree))\n",
        "print('Bias(median value of delta z):', np.median(y_test_tree-y_pred_tree))\n",
        "plt.show()\n",
        "\n",
        "delta_znorm=abs((y_test_tree-y_pred_tree)-np.median(y_pred_tree-y_test_tree)/(1+y_test_tree))\n",
        "NMAD=1.48*np.median(delta_znorm)\n",
        "print('NMAD: ', NMAD)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Catastropic Outliers:  1.6303834070119831 %\n",
            "Total rms:  0    0.025886\n",
            "dtype: float64\n",
            "rms w/o outliers 0    0.013326\n",
            "dtype: float64\n",
            "Bias:  0   -0.003654\n",
            "dtype: float64\n",
            "Bias(median value of delta z): -1.1102230246251565e-16\n",
            "NMAD:  5.510970103487969e-06\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "njsL0QZWCE7n"
      },
      "source": [
        "y_test_tree=scaler.inverse_transform(y_test_tree)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m4-06RxdQTA1"
      },
      "source": [
        "y_pred_tree=scaler.inverse_transform(y_pred_tree)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KsxPMqP9PrBY",
        "outputId": "5a92c762-98cc-42e2-cde7-71a09f7190cb"
      },
      "source": [
        "y_pred_tree"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[6.54006617],\n",
              "       [6.96608661],\n",
              "       [4.17019958],\n",
              "       ...,\n",
              "       [4.867555  ],\n",
              "       [5.970582  ],\n",
              "       [4.061209  ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y-d7PO-WStj_",
        "outputId": "d97ce301-ac66-4eba-f9fc-88601e955b17"
      },
      "source": [
        "y_test_tree"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[6.540075],\n",
              "       [6.966124],\n",
              "       [4.169907],\n",
              "       ...,\n",
              "       [4.867555],\n",
              "       [5.970582],\n",
              "       [4.061209]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 311
        },
        "id": "rOKJC4MgqTXj",
        "outputId": "be29e244-177e-4c06-90bd-769444d96446"
      },
      "source": [
        "import matplotlib.pyplot as plt \n",
        "plt.scatter(y_test_tree[:359670], y_pred_tree[:359670], s=0.4)\n",
        "plt.plot(np.linspace(0,7,7), np.linspace(0,7,7), c = 'orange', linestyle='--')\n",
        "plt.xlim(0, 7)\n",
        "plt.ylim(0, 7)\n",
        "plt.title('True/Predicted Graph') \n",
        "plt.grid(True) \n",
        "plt.xlabel(\"True\") \n",
        "plt.ylabel('Predicted')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Predicted')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        },
        {
          "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": "INDEU0iBSdux"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9qVWx2Ia2pQa",
        "outputId": "1bd36f31-ed91-41e9-f10c-57824a9edeec"
      },
      "source": [
        "!pip install plot_keras_history"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting plot_keras_history\n",
            "  Downloading https://files.pythonhosted.org/packages/73/55/183a10ac8b62ca0411bd93286d394ce65d952e09b8d51d9706a4818afe9c/plot_keras_history-1.1.29.tar.gz\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from plot_keras_history) (3.2.2)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.7/dist-packages (from plot_keras_history) (1.1.5)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from plot_keras_history) (1.4.1)\n",
            "Collecting sanitize_ml_labels\n",
            "  Downloading https://files.pythonhosted.org/packages/57/2c/c97c52a183b3da277a6ca276a9587ea011c5f7bcce0b192d1eaf7eb5084d/sanitize_ml_labels-1.0.26.tar.gz\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->plot_keras_history) (0.10.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->plot_keras_history) (2.4.7)\n",
            "Requirement already satisfied: numpy>=1.11 in /usr/local/lib/python3.7/dist-packages (from matplotlib->plot_keras_history) (1.19.5)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->plot_keras_history) (1.3.1)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->plot_keras_history) (2.8.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas->plot_keras_history) (2018.9)\n",
            "Collecting compress_json\n",
            "  Downloading https://files.pythonhosted.org/packages/c0/9d/1a79dbc3b9d69c57be363b4cc8bd6f278f919a0973f46c7bb831438c9333/compress_json-1.0.4.tar.gz\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from cycler>=0.10->matplotlib->plot_keras_history) (1.15.0)\n",
            "Building wheels for collected packages: plot-keras-history, sanitize-ml-labels, compress-json\n",
            "  Building wheel for plot-keras-history (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for plot-keras-history: filename=plot_keras_history-1.1.29-cp37-none-any.whl size=8614 sha256=cae35e6e44e694f72fe2810c7ef1e48650f44332f77333077084d21cc56399af\n",
            "  Stored in directory: /root/.cache/pip/wheels/8b/cb/fb/cbeba24d72e736b8987b7d0dc37afc0e76f3c16e544ceff727\n",
            "  Building wheel for sanitize-ml-labels (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for sanitize-ml-labels: filename=sanitize_ml_labels-1.0.26-cp37-none-any.whl size=7610 sha256=9c0f4ce1ecc22b260867dd3724e5fa4e37af745cbd1bc95aa72176d087e03e42\n",
            "  Stored in directory: /root/.cache/pip/wheels/2a/62/31/d348320becc60729f508a2c5aad445b94d7543c5cb577cfb9d\n",
            "  Building wheel for compress-json (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for compress-json: filename=compress_json-1.0.4-cp37-none-any.whl size=4586 sha256=9c3da2fe375a7a06d6629ce70bc85125e5ae7d07f56643d030d3833408c2e5fc\n",
            "  Stored in directory: /root/.cache/pip/wheels/83/ec/21/51460dd508e4a152c0643946c21fae84eb3391171164d35745\n",
            "Successfully built plot-keras-history sanitize-ml-labels compress-json\n",
            "Installing collected packages: compress-json, sanitize-ml-labels, plot-keras-history\n",
            "Successfully installed compress-json-1.0.4 plot-keras-history-1.1.29 sanitize-ml-labels-1.0.26\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "MlBzu6592uVm",
        "outputId": "cdb0becb-b973-4695-82e1-723901b0950f"
      },
      "source": [
        "import timeit\n",
        "#from keras.wrappers.scikit_learn import KerasRegressor\n",
        "#from sklearn.model_selection import cross_val_score\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout\n",
        "from plot_keras_history import plot_history\n",
        "from keras.layers import BatchNormalization\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from keras.wrappers.scikit_learn import KerasRegressor\n",
        "\n",
        "#with 1000 batchsize overfitting is much- RMS r2 are not good..256 batchsize was ok\n",
        "\n",
        "model = Sequential()\n",
        "# The Input Layer :\n",
        "model.add(Dense(256, kernel_initializer='he_uniform',input_dim = 25, activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(512, kernel_initializer='he_uniform',activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "# The Hidden Layers :\n",
        "model.add(Dense(912, kernel_initializer='he_uniform',activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(512, kernel_initializer='uniform',activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(256, kernel_initializer='he_uniform',activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(1, kernel_initializer='he_uniform',activation='linear'))\n",
        "opt=keras.optimizers.Adamax(lr=.001)\n",
        "start = timeit.default_timer()\n",
        "model.compile(loss='mean_absolute_error', optimizer=opt, metrics=['mae'])\n",
        "history= model.fit(X_train, y_train, batch_size=256, epochs =200, verbose=1,validation_data=(X_val, y_val)).history\n",
        "model.summary()\n",
        "stop = timeit.default_timer()\n",
        "print('Time: ', (stop - start)/60,'minutes')  \n",
        "plot_history(history)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/200\n",
            "10538/10538 [==============================] - 37s 3ms/step - loss: 0.2144 - mae: 0.2144 - val_loss: 0.1529 - val_mae: 0.1529\n",
            "Epoch 2/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.1570 - mae: 0.1570 - val_loss: 0.1487 - val_mae: 0.1487\n",
            "Epoch 3/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.1537 - mae: 0.1537 - val_loss: 0.1470 - val_mae: 0.1470\n",
            "Epoch 4/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1524 - mae: 0.1524 - val_loss: 0.1458 - val_mae: 0.1458\n",
            "Epoch 5/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1511 - mae: 0.1511 - val_loss: 0.1455 - val_mae: 0.1455\n",
            "Epoch 6/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1499 - mae: 0.1499 - val_loss: 0.1447 - val_mae: 0.1447\n",
            "Epoch 7/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1489 - mae: 0.1489 - val_loss: 0.1414 - val_mae: 0.1414\n",
            "Epoch 8/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1478 - mae: 0.1478 - val_loss: 0.1403 - val_mae: 0.1403\n",
            "Epoch 9/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1469 - mae: 0.1469 - val_loss: 0.1388 - val_mae: 0.1388\n",
            "Epoch 10/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1459 - mae: 0.1459 - val_loss: 0.1381 - val_mae: 0.1381\n",
            "Epoch 11/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1448 - mae: 0.1448 - val_loss: 0.1378 - val_mae: 0.1378\n",
            "Epoch 12/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1444 - mae: 0.1444 - val_loss: 0.1340 - val_mae: 0.1340\n",
            "Epoch 13/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1429 - mae: 0.1429 - val_loss: 0.1330 - val_mae: 0.1330\n",
            "Epoch 14/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1421 - mae: 0.1421 - val_loss: 0.1330 - val_mae: 0.1330\n",
            "Epoch 15/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1409 - mae: 0.1409 - val_loss: 0.1292 - val_mae: 0.1292\n",
            "Epoch 16/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1402 - mae: 0.1402 - val_loss: 0.1292 - val_mae: 0.1292\n",
            "Epoch 17/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1390 - mae: 0.1390 - val_loss: 0.1268 - val_mae: 0.1268\n",
            "Epoch 18/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1383 - mae: 0.1383 - val_loss: 0.1242 - val_mae: 0.1242\n",
            "Epoch 19/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1374 - mae: 0.1374 - val_loss: 0.1273 - val_mae: 0.1273\n",
            "Epoch 20/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1364 - mae: 0.1364 - val_loss: 0.1231 - val_mae: 0.1231\n",
            "Epoch 21/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1357 - mae: 0.1357 - val_loss: 0.1244 - val_mae: 0.1244\n",
            "Epoch 22/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1346 - mae: 0.1346 - val_loss: 0.1236 - val_mae: 0.1236\n",
            "Epoch 23/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1338 - mae: 0.1338 - val_loss: 0.1183 - val_mae: 0.1183\n",
            "Epoch 24/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1328 - mae: 0.1328 - val_loss: 0.1157 - val_mae: 0.1157\n",
            "Epoch 25/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1319 - mae: 0.1319 - val_loss: 0.1208 - val_mae: 0.1208\n",
            "Epoch 26/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1311 - mae: 0.1311 - val_loss: 0.1141 - val_mae: 0.1141\n",
            "Epoch 27/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1303 - mae: 0.1303 - val_loss: 0.1123 - val_mae: 0.1123\n",
            "Epoch 28/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1296 - mae: 0.1296 - val_loss: 0.1115 - val_mae: 0.1115\n",
            "Epoch 29/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1288 - mae: 0.1288 - val_loss: 0.1105 - val_mae: 0.1105\n",
            "Epoch 30/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1281 - mae: 0.1281 - val_loss: 0.1102 - val_mae: 0.1102\n",
            "Epoch 31/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1273 - mae: 0.1273 - val_loss: 0.1075 - val_mae: 0.1075\n",
            "Epoch 32/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1267 - mae: 0.1267 - val_loss: 0.1064 - val_mae: 0.1064\n",
            "Epoch 33/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1254 - mae: 0.1254 - val_loss: 0.1066 - val_mae: 0.1066\n",
            "Epoch 34/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1247 - mae: 0.1247 - val_loss: 0.1046 - val_mae: 0.1046\n",
            "Epoch 35/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1241 - mae: 0.1241 - val_loss: 0.1029 - val_mae: 0.1029\n",
            "Epoch 36/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1234 - mae: 0.1234 - val_loss: 0.1017 - val_mae: 0.1017\n",
            "Epoch 37/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1227 - mae: 0.1227 - val_loss: 0.1025 - val_mae: 0.1025\n",
            "Epoch 38/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1220 - mae: 0.1220 - val_loss: 0.1016 - val_mae: 0.1016\n",
            "Epoch 39/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1214 - mae: 0.1214 - val_loss: 0.0987 - val_mae: 0.0987\n",
            "Epoch 40/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1206 - mae: 0.1206 - val_loss: 0.1039 - val_mae: 0.1039\n",
            "Epoch 41/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1202 - mae: 0.1202 - val_loss: 0.1039 - val_mae: 0.1039\n",
            "Epoch 42/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1195 - mae: 0.1195 - val_loss: 0.0981 - val_mae: 0.0981\n",
            "Epoch 43/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1187 - mae: 0.1187 - val_loss: 0.0975 - val_mae: 0.0975\n",
            "Epoch 44/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1180 - mae: 0.1180 - val_loss: 0.0947 - val_mae: 0.0947\n",
            "Epoch 45/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1177 - mae: 0.1177 - val_loss: 0.0924 - val_mae: 0.0924\n",
            "Epoch 46/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1169 - mae: 0.1169 - val_loss: 0.0936 - val_mae: 0.0936\n",
            "Epoch 47/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1164 - mae: 0.1164 - val_loss: 0.0972 - val_mae: 0.0972\n",
            "Epoch 48/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1156 - mae: 0.1156 - val_loss: 0.0929 - val_mae: 0.0929\n",
            "Epoch 49/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1153 - mae: 0.1153 - val_loss: 0.0909 - val_mae: 0.0909\n",
            "Epoch 50/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1147 - mae: 0.1147 - val_loss: 0.0899 - val_mae: 0.0899\n",
            "Epoch 51/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1145 - mae: 0.1145 - val_loss: 0.0900 - val_mae: 0.0900\n",
            "Epoch 52/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1137 - mae: 0.1137 - val_loss: 0.0894 - val_mae: 0.0894\n",
            "Epoch 53/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1130 - mae: 0.1130 - val_loss: 0.0898 - val_mae: 0.0898\n",
            "Epoch 54/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1124 - mae: 0.1124 - val_loss: 0.0871 - val_mae: 0.0871\n",
            "Epoch 55/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1123 - mae: 0.1123 - val_loss: 0.0886 - val_mae: 0.0886\n",
            "Epoch 56/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1117 - mae: 0.1117 - val_loss: 0.0860 - val_mae: 0.0860\n",
            "Epoch 57/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1113 - mae: 0.1113 - val_loss: 0.0851 - val_mae: 0.0851\n",
            "Epoch 58/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1109 - mae: 0.1109 - val_loss: 0.0843 - val_mae: 0.0843\n",
            "Epoch 59/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1104 - mae: 0.1104 - val_loss: 0.0850 - val_mae: 0.0850\n",
            "Epoch 60/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1100 - mae: 0.1100 - val_loss: 0.0851 - val_mae: 0.0851\n",
            "Epoch 61/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1096 - mae: 0.1096 - val_loss: 0.0862 - val_mae: 0.0862\n",
            "Epoch 62/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.1091 - mae: 0.1091 - val_loss: 0.0842 - val_mae: 0.0842\n",
            "Epoch 63/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1087 - mae: 0.1087 - val_loss: 0.0882 - val_mae: 0.0882\n",
            "Epoch 64/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1084 - mae: 0.1084 - val_loss: 0.0823 - val_mae: 0.0823\n",
            "Epoch 65/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1079 - mae: 0.1079 - val_loss: 0.0815 - val_mae: 0.0815\n",
            "Epoch 66/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1072 - mae: 0.1072 - val_loss: 0.0812 - val_mae: 0.0812\n",
            "Epoch 67/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1070 - mae: 0.1070 - val_loss: 0.0827 - val_mae: 0.0827\n",
            "Epoch 68/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1066 - mae: 0.1066 - val_loss: 0.0841 - val_mae: 0.0841\n",
            "Epoch 69/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1065 - mae: 0.1065 - val_loss: 0.0791 - val_mae: 0.0791\n",
            "Epoch 70/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1063 - mae: 0.1063 - val_loss: 0.0799 - val_mae: 0.0799\n",
            "Epoch 71/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1058 - mae: 0.1058 - val_loss: 0.0791 - val_mae: 0.0791\n",
            "Epoch 72/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1053 - mae: 0.1053 - val_loss: 0.0810 - val_mae: 0.0810\n",
            "Epoch 73/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1050 - mae: 0.1050 - val_loss: 0.0798 - val_mae: 0.0798\n",
            "Epoch 74/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1046 - mae: 0.1046 - val_loss: 0.0790 - val_mae: 0.0790\n",
            "Epoch 75/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.1044 - mae: 0.1044 - val_loss: 0.0788 - val_mae: 0.0788\n",
            "Epoch 76/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1039 - mae: 0.1039 - val_loss: 0.0773 - val_mae: 0.0773\n",
            "Epoch 77/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1037 - mae: 0.1037 - val_loss: 0.0789 - val_mae: 0.0789\n",
            "Epoch 78/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1033 - mae: 0.1033 - val_loss: 0.0787 - val_mae: 0.0787\n",
            "Epoch 79/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1032 - mae: 0.1032 - val_loss: 0.0769 - val_mae: 0.0769\n",
            "Epoch 80/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1028 - mae: 0.1028 - val_loss: 0.0768 - val_mae: 0.0768\n",
            "Epoch 81/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1026 - mae: 0.1026 - val_loss: 0.0764 - val_mae: 0.0764\n",
            "Epoch 82/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1024 - mae: 0.1024 - val_loss: 0.0754 - val_mae: 0.0754\n",
            "Epoch 83/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1018 - mae: 0.1018 - val_loss: 0.0756 - val_mae: 0.0756\n",
            "Epoch 84/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1018 - mae: 0.1018 - val_loss: 0.0769 - val_mae: 0.0769\n",
            "Epoch 85/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1014 - mae: 0.1014 - val_loss: 0.0759 - val_mae: 0.0759\n",
            "Epoch 86/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1010 - mae: 0.1010 - val_loss: 0.0753 - val_mae: 0.0753\n",
            "Epoch 87/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1011 - mae: 0.1011 - val_loss: 0.0752 - val_mae: 0.0752\n",
            "Epoch 88/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1005 - mae: 0.1005 - val_loss: 0.0751 - val_mae: 0.0751\n",
            "Epoch 89/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1004 - mae: 0.1004 - val_loss: 0.0736 - val_mae: 0.0736\n",
            "Epoch 90/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.1003 - mae: 0.1003 - val_loss: 0.0729 - val_mae: 0.0729\n",
            "Epoch 91/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0999 - mae: 0.0999 - val_loss: 0.0743 - val_mae: 0.0743\n",
            "Epoch 92/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0997 - mae: 0.0997 - val_loss: 0.0726 - val_mae: 0.0726\n",
            "Epoch 93/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0995 - mae: 0.0995 - val_loss: 0.0732 - val_mae: 0.0732\n",
            "Epoch 94/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0993 - mae: 0.0993 - val_loss: 0.0723 - val_mae: 0.0723\n",
            "Epoch 95/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0989 - mae: 0.0989 - val_loss: 0.0720 - val_mae: 0.0720\n",
            "Epoch 96/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0990 - mae: 0.0990 - val_loss: 0.0728 - val_mae: 0.0728\n",
            "Epoch 97/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0984 - mae: 0.0984 - val_loss: 0.0706 - val_mae: 0.0706\n",
            "Epoch 98/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0982 - mae: 0.0982 - val_loss: 0.0701 - val_mae: 0.0701\n",
            "Epoch 99/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0981 - mae: 0.0981 - val_loss: 0.0714 - val_mae: 0.0714\n",
            "Epoch 100/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0980 - mae: 0.0980 - val_loss: 0.0725 - val_mae: 0.0725\n",
            "Epoch 101/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0976 - mae: 0.0976 - val_loss: 0.0700 - val_mae: 0.0700\n",
            "Epoch 102/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0976 - mae: 0.0976 - val_loss: 0.0705 - val_mae: 0.0705\n",
            "Epoch 103/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0973 - mae: 0.0973 - val_loss: 0.0701 - val_mae: 0.0701\n",
            "Epoch 104/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0970 - mae: 0.0970 - val_loss: 0.0698 - val_mae: 0.0698\n",
            "Epoch 105/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0967 - mae: 0.0967 - val_loss: 0.0709 - val_mae: 0.0709\n",
            "Epoch 106/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0968 - mae: 0.0968 - val_loss: 0.0690 - val_mae: 0.0690\n",
            "Epoch 107/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0964 - mae: 0.0964 - val_loss: 0.0705 - val_mae: 0.0705\n",
            "Epoch 108/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0962 - mae: 0.0962 - val_loss: 0.0692 - val_mae: 0.0692\n",
            "Epoch 109/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0961 - mae: 0.0961 - val_loss: 0.0692 - val_mae: 0.0692\n",
            "Epoch 110/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0959 - mae: 0.0959 - val_loss: 0.0705 - val_mae: 0.0705\n",
            "Epoch 111/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0958 - mae: 0.0958 - val_loss: 0.0679 - val_mae: 0.0679\n",
            "Epoch 112/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.0954 - mae: 0.0954 - val_loss: 0.0687 - val_mae: 0.0687\n",
            "Epoch 113/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0954 - mae: 0.0954 - val_loss: 0.0688 - val_mae: 0.0688\n",
            "Epoch 114/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0951 - mae: 0.0951 - val_loss: 0.0663 - val_mae: 0.0663\n",
            "Epoch 115/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0948 - mae: 0.0948 - val_loss: 0.0677 - val_mae: 0.0677\n",
            "Epoch 116/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0946 - mae: 0.0946 - val_loss: 0.0666 - val_mae: 0.0666\n",
            "Epoch 117/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0945 - mae: 0.0945 - val_loss: 0.0666 - val_mae: 0.0666\n",
            "Epoch 118/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0943 - mae: 0.0943 - val_loss: 0.0663 - val_mae: 0.0663\n",
            "Epoch 119/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0941 - mae: 0.0941 - val_loss: 0.0681 - val_mae: 0.0681\n",
            "Epoch 120/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0939 - mae: 0.0939 - val_loss: 0.0674 - val_mae: 0.0674\n",
            "Epoch 121/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0936 - mae: 0.0936 - val_loss: 0.0655 - val_mae: 0.0655\n",
            "Epoch 122/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0934 - mae: 0.0934 - val_loss: 0.0667 - val_mae: 0.0667\n",
            "Epoch 123/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0938 - mae: 0.0938 - val_loss: 0.0692 - val_mae: 0.0692\n",
            "Epoch 124/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0934 - mae: 0.0934 - val_loss: 0.0659 - val_mae: 0.0659\n",
            "Epoch 125/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0933 - mae: 0.0933 - val_loss: 0.0671 - val_mae: 0.0671\n",
            "Epoch 126/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0930 - mae: 0.0930 - val_loss: 0.0646 - val_mae: 0.0646\n",
            "Epoch 127/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0931 - mae: 0.0931 - val_loss: 0.0671 - val_mae: 0.0671\n",
            "Epoch 128/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0929 - mae: 0.0929 - val_loss: 0.0659 - val_mae: 0.0659\n",
            "Epoch 129/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0929 - mae: 0.0929 - val_loss: 0.0641 - val_mae: 0.0641\n",
            "Epoch 130/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0924 - mae: 0.0924 - val_loss: 0.0655 - val_mae: 0.0655\n",
            "Epoch 131/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0923 - mae: 0.0923 - val_loss: 0.0645 - val_mae: 0.0645\n",
            "Epoch 132/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0921 - mae: 0.0921 - val_loss: 0.0649 - val_mae: 0.0649\n",
            "Epoch 133/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0921 - mae: 0.0921 - val_loss: 0.0641 - val_mae: 0.0641\n",
            "Epoch 134/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.0919 - mae: 0.0919 - val_loss: 0.0630 - val_mae: 0.0630\n",
            "Epoch 135/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0918 - mae: 0.0918 - val_loss: 0.0637 - val_mae: 0.0637\n",
            "Epoch 136/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0918 - mae: 0.0918 - val_loss: 0.0633 - val_mae: 0.0633\n",
            "Epoch 137/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0915 - mae: 0.0915 - val_loss: 0.0644 - val_mae: 0.0644\n",
            "Epoch 138/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0913 - mae: 0.0913 - val_loss: 0.0648 - val_mae: 0.0648\n",
            "Epoch 139/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.0912 - mae: 0.0912 - val_loss: 0.0635 - val_mae: 0.0635\n",
            "Epoch 140/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0911 - mae: 0.0911 - val_loss: 0.0633 - val_mae: 0.0633\n",
            "Epoch 141/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0912 - mae: 0.0912 - val_loss: 0.0634 - val_mae: 0.0634\n",
            "Epoch 142/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0910 - mae: 0.0910 - val_loss: 0.0682 - val_mae: 0.0682\n",
            "Epoch 143/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0908 - mae: 0.0908 - val_loss: 0.0635 - val_mae: 0.0635\n",
            "Epoch 144/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0905 - mae: 0.0905 - val_loss: 0.0639 - val_mae: 0.0639\n",
            "Epoch 145/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0903 - mae: 0.0903 - val_loss: 0.0645 - val_mae: 0.0645\n",
            "Epoch 146/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0902 - mae: 0.0902 - val_loss: 0.0628 - val_mae: 0.0628\n",
            "Epoch 147/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0902 - mae: 0.0902 - val_loss: 0.0615 - val_mae: 0.0615\n",
            "Epoch 148/200\n",
            "10538/10538 [==============================] - 31s 3ms/step - loss: 0.0900 - mae: 0.0900 - val_loss: 0.0627 - val_mae: 0.0627\n",
            "Epoch 149/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0900 - mae: 0.0900 - val_loss: 0.0618 - val_mae: 0.0618\n",
            "Epoch 150/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0897 - mae: 0.0897 - val_loss: 0.0629 - val_mae: 0.0629\n",
            "Epoch 151/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0899 - mae: 0.0899 - val_loss: 0.0611 - val_mae: 0.0611\n",
            "Epoch 152/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0897 - mae: 0.0897 - val_loss: 0.0623 - val_mae: 0.0623\n",
            "Epoch 153/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0895 - mae: 0.0895 - val_loss: 0.0626 - val_mae: 0.0626\n",
            "Epoch 154/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0893 - mae: 0.0893 - val_loss: 0.0628 - val_mae: 0.0628\n",
            "Epoch 155/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0892 - mae: 0.0892 - val_loss: 0.0607 - val_mae: 0.0607\n",
            "Epoch 156/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0892 - mae: 0.0892 - val_loss: 0.0605 - val_mae: 0.0605\n",
            "Epoch 157/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0890 - mae: 0.0890 - val_loss: 0.0627 - val_mae: 0.0627\n",
            "Epoch 158/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0890 - mae: 0.0890 - val_loss: 0.0621 - val_mae: 0.0621\n",
            "Epoch 159/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0889 - mae: 0.0889 - val_loss: 0.0638 - val_mae: 0.0638\n",
            "Epoch 160/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0887 - mae: 0.0887 - val_loss: 0.0625 - val_mae: 0.0625\n",
            "Epoch 161/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0889 - mae: 0.0889 - val_loss: 0.0612 - val_mae: 0.0612\n",
            "Epoch 162/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0887 - mae: 0.0887 - val_loss: 0.0604 - val_mae: 0.0604\n",
            "Epoch 163/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0886 - mae: 0.0886 - val_loss: 0.0607 - val_mae: 0.0607\n",
            "Epoch 164/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0884 - mae: 0.0884 - val_loss: 0.0644 - val_mae: 0.0644\n",
            "Epoch 165/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0885 - mae: 0.0885 - val_loss: 0.0604 - val_mae: 0.0604\n",
            "Epoch 166/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0881 - mae: 0.0881 - val_loss: 0.0598 - val_mae: 0.0598\n",
            "Epoch 167/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0882 - mae: 0.0882 - val_loss: 0.0603 - val_mae: 0.0603\n",
            "Epoch 168/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0880 - mae: 0.0880 - val_loss: 0.0594 - val_mae: 0.0594\n",
            "Epoch 169/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0878 - mae: 0.0878 - val_loss: 0.0625 - val_mae: 0.0625\n",
            "Epoch 170/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0880 - mae: 0.0880 - val_loss: 0.0599 - val_mae: 0.0599\n",
            "Epoch 171/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0878 - mae: 0.0878 - val_loss: 0.0600 - val_mae: 0.0600\n",
            "Epoch 172/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0875 - mae: 0.0875 - val_loss: 0.0621 - val_mae: 0.0621\n",
            "Epoch 173/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0876 - mae: 0.0876 - val_loss: 0.0630 - val_mae: 0.0630\n",
            "Epoch 174/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0876 - mae: 0.0876 - val_loss: 0.0594 - val_mae: 0.0594\n",
            "Epoch 175/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0874 - mae: 0.0874 - val_loss: 0.0594 - val_mae: 0.0594\n",
            "Epoch 176/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0873 - mae: 0.0873 - val_loss: 0.0611 - val_mae: 0.0611\n",
            "Epoch 177/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0873 - mae: 0.0873 - val_loss: 0.0616 - val_mae: 0.0616\n",
            "Epoch 178/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0872 - mae: 0.0872 - val_loss: 0.0594 - val_mae: 0.0594\n",
            "Epoch 179/200\n",
            "10538/10538 [==============================] - 34s 3ms/step - loss: 0.0871 - mae: 0.0871 - val_loss: 0.0606 - val_mae: 0.0606\n",
            "Epoch 180/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0871 - mae: 0.0871 - val_loss: 0.0589 - val_mae: 0.0589\n",
            "Epoch 181/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0868 - mae: 0.0868 - val_loss: 0.0646 - val_mae: 0.0646\n",
            "Epoch 182/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0870 - mae: 0.0870 - val_loss: 0.0592 - val_mae: 0.0592\n",
            "Epoch 183/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0868 - mae: 0.0868 - val_loss: 0.0584 - val_mae: 0.0584\n",
            "Epoch 184/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0863 - mae: 0.0863 - val_loss: 0.0580 - val_mae: 0.0580\n",
            "Epoch 185/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0867 - mae: 0.0867 - val_loss: 0.0589 - val_mae: 0.0589\n",
            "Epoch 186/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0865 - mae: 0.0865 - val_loss: 0.0606 - val_mae: 0.0606\n",
            "Epoch 187/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0865 - mae: 0.0865 - val_loss: 0.0604 - val_mae: 0.0604\n",
            "Epoch 188/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0863 - mae: 0.0863 - val_loss: 0.0601 - val_mae: 0.0601\n",
            "Epoch 189/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0865 - mae: 0.0865 - val_loss: 0.0591 - val_mae: 0.0591\n",
            "Epoch 190/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0863 - mae: 0.0863 - val_loss: 0.0576 - val_mae: 0.0576\n",
            "Epoch 191/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0862 - mae: 0.0862 - val_loss: 0.0594 - val_mae: 0.0594\n",
            "Epoch 192/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0861 - mae: 0.0861 - val_loss: 0.0577 - val_mae: 0.0577\n",
            "Epoch 193/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0858 - mae: 0.0858 - val_loss: 0.0591 - val_mae: 0.0591\n",
            "Epoch 194/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0858 - mae: 0.0858 - val_loss: 0.0609 - val_mae: 0.0609\n",
            "Epoch 195/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0859 - mae: 0.0859 - val_loss: 0.0600 - val_mae: 0.0600\n",
            "Epoch 196/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0857 - mae: 0.0857 - val_loss: 0.0581 - val_mae: 0.0581\n",
            "Epoch 197/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0856 - mae: 0.0856 - val_loss: 0.0589 - val_mae: 0.0589\n",
            "Epoch 198/200\n",
            "10538/10538 [==============================] - 33s 3ms/step - loss: 0.0855 - mae: 0.0855 - val_loss: 0.0584 - val_mae: 0.0584\n",
            "Epoch 199/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0854 - mae: 0.0854 - val_loss: 0.0578 - val_mae: 0.0578\n",
            "Epoch 200/200\n",
            "10538/10538 [==============================] - 32s 3ms/step - loss: 0.0854 - mae: 0.0854 - val_loss: 0.0575 - val_mae: 0.0575\n",
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense (Dense)                (None, 256)               6656      \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 512)               131584    \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 912)               467856    \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 912)               0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 512)               467456    \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 256)               131328    \n",
            "_________________________________________________________________\n",
            "dropout_4 (Dropout)          (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 1)                 257       \n",
            "=================================================================\n",
            "Total params: 1,205,137\n",
            "Trainable params: 1,205,137\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Time:  106.18392263391667 minutes\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<Figure size 720x360 with 2 Axes>,\n",
              " array([<matplotlib.axes._subplots.AxesSubplot object at 0x7fd226cd1e50>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fd226ca3850>],\n",
              "       dtype=object))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yVRG6Eqe21Ma",
        "outputId": "bdf5ffa6-2ad4-4fd3-d771-de07eb6e3bd8"
      },
      "source": [
        "import numpy as geek \n",
        "y_pred_ann=model.predict(X_test)\n",
        "#print(y_pred)\n",
        "out_arr = geek.column_stack((y_pred_ann, y_test)) #comparing two values\n",
        "print (\"Output stacked array:\\n \", out_arr) "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Output stacked array:\n",
            "  [[0.95972538 0.93430831]\n",
            " [0.97964358 0.99517322]\n",
            " [0.58917129 0.59570857]\n",
            " ...\n",
            " [0.69577467 0.69537384]\n",
            " [0.79759699 0.85295113]\n",
            " [0.54268479 0.58018009]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MBhny50zDnCU",
        "outputId": "964a218c-5336-4e73-981a-4870d8e25ac0"
      },
      "source": [
        "from sklearn import metrics\n",
        "from math import sqrt\n",
        "from sklearn.metrics import mean_squared_error\n",
        "var = metrics.explained_variance_score(y_test, y_pred_ann)\n",
        "mae = metrics.mean_absolute_error(y_test, y_pred_ann)\n",
        "mse = metrics.mean_squared_error(y_test, y_pred_ann)\n",
        "r2 = metrics.r2_score(y_test, y_pred_ann)\n",
        "rmse = sqrt(mean_squared_error(y_test, y_pred_ann))\n",
        "print ('Explained variance (best possible score is 1.0, lower values are worse):', var)\n",
        "print ('Mean Absolute Error (best is 0.0):', mae)\n",
        "print ('Mean Squared Error (best is 0.0):', mse)\n",
        "print ('R2 score (best is 1.0):', r2)\n",
        "print ('root Mean Squared Error (best is 0.0):', rmse)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Explained variance (best possible score is 1.0, lower values are worse): 0.7743113978177752\n",
            "Mean Absolute Error (best is 0.0): 0.05716066432142311\n",
            "Mean Squared Error (best is 0.0): 0.019627651296268322\n",
            "R2 score (best is 1.0): 0.77143753145708\n",
            "root Mean Squared Error (best is 0.0): 0.14009871982380254\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fYbDBfs8vOyi",
        "outputId": "1be3a480-4aa4-4418-b184-36e90d154925"
      },
      "source": [
        "y_test_ann=y_test\n",
        "y_test_ann=pd.DataFrame(y_test_ann)\n",
        "y_pred_ann=pd.DataFrame(y_pred_ann)\n",
        "outl=y_pred_ann[abs(y_test-y_pred_ann)>0.1]\n",
        "outl=outl.dropna()\n",
        "print('Catastropic Outliers: ', outl.shape[0]*100.0/y_test.shape[0],'%')\n",
        "print('Total rms: ', np.sqrt(np.mean((y_test_ann-y_pred_ann)**2)))\n",
        "print('rms w/o outliers', np.sqrt(np.mean(((y_test_ann-y_pred_ann)[abs(y_test_ann-y_pred_ann)<0.1])**2)))\n",
        "print('Bias: ', np.mean(y_test_ann-y_pred_ann))\n",
        "print('Bias(median value of delta z):', np.median(y_test_ann-y_pred_ann))\n",
        "plt.show()\n",
        "\n",
        "delta_znorm=abs((y_test_ann-y_pred_ann)-np.median(y_pred_ann-y_test_ann)/(1+y_test_ann))\n",
        "NMAD=1.48*np.median(delta_znorm)\n",
        "print('NMAD: ', NMAD)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Catastropic Outliers:  12.011288125225901 %\n",
            "Total rms:  0    0.140099\n",
            "dtype: float64\n",
            "rms w/o outliers 0    0.026571\n",
            "dtype: float64\n",
            "Bias:  0   -0.01571\n",
            "dtype: float64\n",
            "Bias(median value of delta z): -0.004164767973104604\n",
            "NMAD:  0.023316773073038793\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UzDKMvSS5Lxm"
      },
      "source": [
        "y_test_ann=scaler.inverse_transform(y_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dGejsRMcTqnN"
      },
      "source": [
        "y_pred_ann=scaler.inverse_transform(y_pred_ann)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 311
        },
        "id": "H1Y-OVrjebJ5",
        "outputId": "ef943011-2225-4b19-949e-04115cf1b244"
      },
      "source": [
        "import matplotlib.pyplot as plt \n",
        "plt.scatter(y_test_ann[:150000], y_pred_ann[:150000], s=0.4)\n",
        "plt.plot(np.linspace(0,7,7), np.linspace(0,7,7), c = 'orange', linestyle='--')\n",
        "plt.xlim(0, 7)\n",
        "plt.ylim(0, 7)\n",
        "plt.title('True/Predicted Graph') \n",
        "plt.grid(True) \n",
        "plt.xlabel(\"True\") \n",
        "plt.ylabel('Predicted')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0, 0.5, 'Predicted')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        },
        {
          "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"
          }
        }
      ]
    }
  ]
}