EMTAB7606_scRNAseqTimeSeries.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aVyxmKdciHHu"
   },
   "source": [
    "# Seeding the environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "executionInfo": {
     "elapsed": 449,
     "status": "ok",
     "timestamp": 1647771997834,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "Dk9waMGV5u9t"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ['TF_CPP_MIN_LOG_LEVEL']='2'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "executionInfo": {
     "elapsed": 7759,
     "status": "ok",
     "timestamp": 1647772005590,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "WbszUtlH5u9w"
   },
   "outputs": [],
   "source": [
    "# Set seed value\n",
    "seed_value = 56\n",
    "import os\n",
    "os.environ['PYTHONHASHSEED']=str(seed_value)\n",
    "# 2. Set `python` built-in pseudo-random generator at a fixed value\n",
    "import random\n",
    "random.seed(seed_value)\n",
    "# 3. Set `numpy` pseudo-random generator at a fixed value\n",
    "import numpy as np\n",
    "np.random.seed(seed_value)\n",
    "# 4. Set `tensorflow` pseudo-random generator at a fixed value\n",
    "import tensorflow as tf\n",
    "tf.random.set_seed(seed_value)\n",
    "# 5. Configure a new global `tensorflow` session\n",
    "from keras import backend as K\n",
    "session_conf = tf.compat.v1.ConfigProto(intra_op_parallelism_threads=1, inter_op_parallelism_threads=1)\n",
    "sess = tf.compat.v1.Session(graph=tf.compat.v1.get_default_graph(), config=session_conf)\n",
    "tf.compat.v1.keras.backend.set_session(sess)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "3mGPjlWJ5u90"
   },
   "source": [
    "# Importing libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "executionInfo": {
     "elapsed": 18,
     "status": "ok",
     "timestamp": 1647772005591,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "g9GvpQWtMSt8"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy.io import mmread\n",
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "C1Vs4mp1iMLL"
   },
   "source": [
    "# Importing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "executionInfo": {
     "elapsed": 15,
     "status": "ok",
     "timestamp": 1647772010506,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "MGg7vJ-55u93"
   },
   "outputs": [],
   "source": [
    "dir = '/home/DibyenduBSeal/DBS_HPC/scRNAseqTimeSeries/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1562,
     "status": "ok",
     "timestamp": 1647772012057,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "cbYYxgs2MYAU",
    "outputId": "ea941660-8ee5-4e30-aa57-22d8e16081df"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<14162x209 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 345731 stored elements in COOrdinate format>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = mmread(dir+'Data/E-MTAB-7606/E-MTAB-7606.aggregated_filtered_normalised_counts.mtx')\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 15,
     "status": "ok",
     "timestamp": 1647772012057,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "vtrkIbcwQqyB",
    "outputId": "0a4c7c20-a78f-4eca-f119-26637d3dbc3a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11.680661707419748"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sparsity=34573100/(14162*209)\n",
    "sparsity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HnKRW6_oiQ1s"
   },
   "source": [
    "# Pre-processing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1kzpHC5-5u9_"
   },
   "source": [
    "Converting sparse matrix into dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 13,
     "status": "ok",
     "timestamp": 1647772012058,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "sZ5yKoiGMawQ",
    "outputId": "73fdc9ea-c9d1-4f6c-b094-eafc39d14908"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
       "        0.       ],\n",
       "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
       "        0.       ],\n",
       "       [0.       , 0.       , 1.7104279, ..., 0.       , 0.       ,\n",
       "        5.4015794],\n",
       "       ...,\n",
       "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
       "        0.       ],\n",
       "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
       "        0.       ],\n",
       "       [0.       , 0.       , 0.       , ..., 0.       , 0.       ,\n",
       "        0.       ]])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_arr = X.toarray()\n",
    "X_arr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 10,
     "status": "ok",
     "timestamp": 1647772012059,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "riCrj5PdG8u3",
    "outputId": "78b786f7-ccdc-43f7-caf7-1e700e534f55"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14162, 209)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_arr.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 393
    },
    "executionInfo": {
     "elapsed": 1577,
     "status": "ok",
     "timestamp": 1647772013630,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "m4SqHpRoz9VB",
    "outputId": "68d78432-fff3-4d35-d342-5c6bb0c7ed02"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>199</th>\n",
       "      <th>200</th>\n",
       "      <th>201</th>\n",
       "      <th>202</th>\n",
       "      <th>203</th>\n",
       "      <th>204</th>\n",
       "      <th>205</th>\n",
       "      <th>206</th>\n",
       "      <th>207</th>\n",
       "      <th>208</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "      <td>14162.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>70.611489</td>\n",
       "      <td>70.611498</td>\n",
       "      <td>70.611493</td>\n",
       "      <td>70.611497</td>\n",
       "      <td>70.611501</td>\n",
       "      <td>70.611497</td>\n",
       "      <td>70.611500</td>\n",
       "      <td>70.611496</td>\n",
       "      <td>70.611497</td>\n",
       "      <td>70.611499</td>\n",
       "      <td>...</td>\n",
       "      <td>70.611494</td>\n",
       "      <td>70.611501</td>\n",
       "      <td>70.611488</td>\n",
       "      <td>70.611494</td>\n",
       "      <td>70.611498</td>\n",
       "      <td>70.611495</td>\n",
       "      <td>70.611491</td>\n",
       "      <td>70.611486</td>\n",
       "      <td>70.611498</td>\n",
       "      <td>70.611501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4744.890451</td>\n",
       "      <td>2894.741937</td>\n",
       "      <td>3518.732531</td>\n",
       "      <td>3508.123727</td>\n",
       "      <td>2922.907360</td>\n",
       "      <td>3273.435815</td>\n",
       "      <td>2111.971493</td>\n",
       "      <td>3685.592832</td>\n",
       "      <td>3868.485084</td>\n",
       "      <td>3125.603025</td>\n",
       "      <td>...</td>\n",
       "      <td>1778.705686</td>\n",
       "      <td>2187.270133</td>\n",
       "      <td>3268.808364</td>\n",
       "      <td>2920.830656</td>\n",
       "      <td>3104.513272</td>\n",
       "      <td>2515.790768</td>\n",
       "      <td>2303.969464</td>\n",
       "      <td>5156.084937</td>\n",
       "      <td>1616.639463</td>\n",
       "      <td>2486.489006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.259066</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>563922.800000</td>\n",
       "      <td>341937.500000</td>\n",
       "      <td>417359.780000</td>\n",
       "      <td>415125.940000</td>\n",
       "      <td>343945.220000</td>\n",
       "      <td>387831.280000</td>\n",
       "      <td>241301.620000</td>\n",
       "      <td>436680.840000</td>\n",
       "      <td>459208.440000</td>\n",
       "      <td>370097.060000</td>\n",
       "      <td>...</td>\n",
       "      <td>202379.860000</td>\n",
       "      <td>255374.200000</td>\n",
       "      <td>386882.440000</td>\n",
       "      <td>344730.440000</td>\n",
       "      <td>367201.780000</td>\n",
       "      <td>293537.030000</td>\n",
       "      <td>269465.160000</td>\n",
       "      <td>613107.100000</td>\n",
       "      <td>183258.280000</td>\n",
       "      <td>291504.340000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 209 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0              1              2              3    \\\n",
       "count   14162.000000   14162.000000   14162.000000   14162.000000   \n",
       "mean       70.611489      70.611498      70.611493      70.611497   \n",
       "std      4744.890451    2894.741937    3518.732531    3508.123727   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         0.000000       0.000000       0.000000       0.000000   \n",
       "max    563922.800000  341937.500000  417359.780000  415125.940000   \n",
       "\n",
       "                 4              5              6              7    \\\n",
       "count   14162.000000   14162.000000   14162.000000   14162.000000   \n",
       "mean       70.611501      70.611497      70.611500      70.611496   \n",
       "std      2922.907360    3273.435815    2111.971493    3685.592832   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         0.000000       0.000000       4.259066       0.000000   \n",
       "max    343945.220000  387831.280000  241301.620000  436680.840000   \n",
       "\n",
       "                 8              9    ...            199            200  \\\n",
       "count   14162.000000   14162.000000  ...   14162.000000   14162.000000   \n",
       "mean       70.611497      70.611499  ...      70.611494      70.611501   \n",
       "std      3868.485084    3125.603025  ...    1778.705686    2187.270133   \n",
       "min         0.000000       0.000000  ...       0.000000       0.000000   \n",
       "25%         0.000000       0.000000  ...       0.000000       0.000000   \n",
       "50%         0.000000       0.000000  ...       0.000000       0.000000   \n",
       "75%         0.000000       0.000000  ...       0.000000       0.000000   \n",
       "max    459208.440000  370097.060000  ...  202379.860000  255374.200000   \n",
       "\n",
       "                 201            202            203            204  \\\n",
       "count   14162.000000   14162.000000   14162.000000   14162.000000   \n",
       "mean       70.611488      70.611494      70.611498      70.611495   \n",
       "std      3268.808364    2920.830656    3104.513272    2515.790768   \n",
       "min         0.000000       0.000000       0.000000       0.000000   \n",
       "25%         0.000000       0.000000       0.000000       0.000000   \n",
       "50%         0.000000       0.000000       0.000000       0.000000   \n",
       "75%         0.000000       0.000000       0.000000       0.000000   \n",
       "max    386882.440000  344730.440000  367201.780000  293537.030000   \n",
       "\n",
       "                 205            206            207            208  \n",
       "count   14162.000000   14162.000000   14162.000000   14162.000000  \n",
       "mean       70.611491      70.611486      70.611498      70.611501  \n",
       "std      2303.969464    5156.084937    1616.639463    2486.489006  \n",
       "min         0.000000       0.000000       0.000000       0.000000  \n",
       "25%         0.000000       0.000000       0.000000       0.000000  \n",
       "50%         0.000000       0.000000       0.000000       0.000000  \n",
       "75%         0.000000       0.000000       0.000000       0.000000  \n",
       "max    269465.160000  613107.100000  183258.280000  291504.340000  \n",
       "\n",
       "[8 rows x 209 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(X_arr).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hRs6748LiXbB"
   },
   "source": [
    "Scaling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "executionInfo": {
     "elapsed": 24,
     "status": "ok",
     "timestamp": 1647772013631,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "BwNbDW3kMeVl",
    "outputId": "a7936d6e-56b1-4ff0-f3e0-8f245b4a39a6"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>14152</th>\n",
       "      <th>14153</th>\n",
       "      <th>14154</th>\n",
       "      <th>14155</th>\n",
       "      <th>14156</th>\n",
       "      <th>14157</th>\n",
       "      <th>14158</th>\n",
       "      <th>14159</th>\n",
       "      <th>14160</th>\n",
       "      <th>14161</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.179058</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.243321</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002232</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.238021</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.032667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 14162 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   0      1         2      3      4      5      6      7      8         9      \\\n",
       "0    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0    0.0  0.179058   \n",
       "1    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0    0.0  0.000000   \n",
       "2    0.0    0.0  0.002232    0.0    0.0    0.0    0.0    0.0    0.0  0.068223   \n",
       "3    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0    0.0  0.000000   \n",
       "4    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0    0.0  0.032667   \n",
       "\n",
       "   ...     14152  14153  14154     14155  14156  14157  14158  14159  14160  \\\n",
       "0  ...  0.000000    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0   \n",
       "1  ...  0.243321    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0   \n",
       "2  ...  0.000000    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0   \n",
       "3  ...  0.000000    0.0    0.0  0.238021    0.0    0.0    0.0    0.0    0.0   \n",
       "4  ...  0.000000    0.0    0.0  0.000000    0.0    0.0    0.0    0.0    0.0   \n",
       "\n",
       "   14161  \n",
       "0    0.0  \n",
       "1    0.0  \n",
       "2    0.0  \n",
       "3    0.0  \n",
       "4    0.0  \n",
       "\n",
       "[5 rows x 14162 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler = MinMaxScaler()\n",
    "X_arr1=scaler.fit_transform(X_arr.T)\n",
    "data = pd.DataFrame(X_arr1)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YxBfSnAAifCg"
   },
   "source": [
    "Extracting row names and column names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 21,
     "status": "ok",
     "timestamp": 1647772013632,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "E6OU8Ru1MlMW",
    "outputId": "423f75d6-78e9-4c36-8302-0f92cf6d2e90"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              0\n",
      "0  SAMEA5215575\n",
      "1  SAMEA5215576\n",
      "2  SAMEA5215577\n",
      "3  SAMEA5215578\n",
      "4  SAMEA5215579\n",
      "                 0\n",
      "0  ENSG00000000003\n",
      "1  ENSG00000000419\n",
      "2  ENSG00000000457\n",
      "3  ENSG00000000460\n",
      "4  ENSG00000000938\n"
     ]
    }
   ],
   "source": [
    "row_names = pd.read_csv(dir+'Data/E-MTAB-7606/E-MTAB-7606.aggregated_filtered_normalised_counts_rows.csv',header=None)\n",
    "col_names = pd.read_csv(dir+'Data/E-MTAB-7606/E-MTAB-7606.aggregated_filtered_normalised_counts_cols.csv',header=None)\n",
    "\n",
    "print(row_names.head())\n",
    "print(col_names.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "executionInfo": {
     "elapsed": 17,
     "status": "ok",
     "timestamp": 1647772013632,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "OeKbmya7MoeO",
    "outputId": "591a097d-3d61-4a48-d239-c8385aac9b55"
   },
   "outputs": [
    {
     "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>ENSG00000000003</th>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <th>ENSG00000000938</th>\n",
       "      <th>ENSG00000000971</th>\n",
       "      <th>ENSG00000001036</th>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <th>ENSG00000001167</th>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <th>...</th>\n",
       "      <th>ENSG00000287846</th>\n",
       "      <th>ENSG00000287856</th>\n",
       "      <th>ENSG00000287924</th>\n",
       "      <th>ENSG00000287977</th>\n",
       "      <th>ENSG00000288218</th>\n",
       "      <th>ENSG00000288258</th>\n",
       "      <th>ENSG00000288380</th>\n",
       "      <th>ENSG00000288525</th>\n",
       "      <th>ENSG00000288534</th>\n",
       "      <th>ENSG00000288579</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SAMEA5215575</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.179058</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SAMEA5215576</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.243321</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SAMEA5215577</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002232</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068223</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SAMEA5215578</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.238021</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SAMEA5215579</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.032667</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 14162 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              ENSG00000000003  ENSG00000000419  ENSG00000000457  \\\n",
       "SAMEA5215575              0.0              0.0         0.000000   \n",
       "SAMEA5215576              0.0              0.0         0.000000   \n",
       "SAMEA5215577              0.0              0.0         0.002232   \n",
       "SAMEA5215578              0.0              0.0         0.000000   \n",
       "SAMEA5215579              0.0              0.0         0.000000   \n",
       "\n",
       "              ENSG00000000460  ENSG00000000938  ENSG00000000971  \\\n",
       "SAMEA5215575              0.0              0.0              0.0   \n",
       "SAMEA5215576              0.0              0.0              0.0   \n",
       "SAMEA5215577              0.0              0.0              0.0   \n",
       "SAMEA5215578              0.0              0.0              0.0   \n",
       "SAMEA5215579              0.0              0.0              0.0   \n",
       "\n",
       "              ENSG00000001036  ENSG00000001084  ENSG00000001167  \\\n",
       "SAMEA5215575              0.0              0.0              0.0   \n",
       "SAMEA5215576              0.0              0.0              0.0   \n",
       "SAMEA5215577              0.0              0.0              0.0   \n",
       "SAMEA5215578              0.0              0.0              0.0   \n",
       "SAMEA5215579              0.0              0.0              0.0   \n",
       "\n",
       "              ENSG00000001460  ...  ENSG00000287846  ENSG00000287856  \\\n",
       "SAMEA5215575         0.179058  ...         0.000000              0.0   \n",
       "SAMEA5215576         0.000000  ...         0.243321              0.0   \n",
       "SAMEA5215577         0.068223  ...         0.000000              0.0   \n",
       "SAMEA5215578         0.000000  ...         0.000000              0.0   \n",
       "SAMEA5215579         0.032667  ...         0.000000              0.0   \n",
       "\n",
       "              ENSG00000287924  ENSG00000287977  ENSG00000288218  \\\n",
       "SAMEA5215575              0.0         0.000000              0.0   \n",
       "SAMEA5215576              0.0         0.000000              0.0   \n",
       "SAMEA5215577              0.0         0.000000              0.0   \n",
       "SAMEA5215578              0.0         0.238021              0.0   \n",
       "SAMEA5215579              0.0         0.000000              0.0   \n",
       "\n",
       "              ENSG00000288258  ENSG00000288380  ENSG00000288525  \\\n",
       "SAMEA5215575              0.0              0.0              0.0   \n",
       "SAMEA5215576              0.0              0.0              0.0   \n",
       "SAMEA5215577              0.0              0.0              0.0   \n",
       "SAMEA5215578              0.0              0.0              0.0   \n",
       "SAMEA5215579              0.0              0.0              0.0   \n",
       "\n",
       "              ENSG00000288534  ENSG00000288579  \n",
       "SAMEA5215575              0.0              0.0  \n",
       "SAMEA5215576              0.0              0.0  \n",
       "SAMEA5215577              0.0              0.0  \n",
       "SAMEA5215578              0.0              0.0  \n",
       "SAMEA5215579              0.0              0.0  \n",
       "\n",
       "[5 rows x 14162 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns=list(col_names[0])\n",
    "data.index=list(row_names[0])\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LEWPqGwScxJx"
   },
   "source": [
    "Removing lowly expressed genes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "executionInfo": {
     "elapsed": 16,
     "status": "ok",
     "timestamp": 1647772013633,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "-i8yBPNbeB3g"
   },
   "outputs": [],
   "source": [
    "tr_data = data.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 17,
     "status": "ok",
     "timestamp": 1647772013634,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "o0wuP4tUcwMV",
    "outputId": "5ab0ef84-6b93-483c-c802-64912707689f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.2549308075756339\n",
      "0.18887653618178296\n",
      "0.1791823570331146\n",
      "0.17141952297643362\n",
      "1.0\n"
     ]
    }
   ],
   "source": [
    "#Sample Example Check\n",
    "for i in tr_data.loc['ENSG00000000003']:\n",
    "  if i>0:\n",
    "    print(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "executionInfo": {
     "elapsed": 3720,
     "status": "ok",
     "timestamp": 1647772017341,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "HHCta5PUdMGW"
   },
   "outputs": [],
   "source": [
    "length=[]\n",
    "for i in tr_data.index:\n",
    "  gen_val=[]\n",
    "  for j in tr_data.loc[i]:\n",
    "    if j>0:\n",
    "      gen_val.append(j)\n",
    "  length.append(len(gen_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "executionInfo": {
     "elapsed": 13,
     "status": "ok",
     "timestamp": 1647772017346,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "pF-TDU55ePMm"
   },
   "outputs": [],
   "source": [
    "amount = [i/tr_data.shape[1] for i in length]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "executionInfo": {
     "elapsed": 13,
     "status": "ok",
     "timestamp": 1647772017347,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "fUj4PjgfeYKv",
    "outputId": "361871d7-5d39-4df7-bcee-dc4bdf531fdb"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>14162.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.116796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.142828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.014354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.033493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.071770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.138756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.995215</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0\n",
       "count  14162.000000\n",
       "mean       0.116796\n",
       "std        0.142828\n",
       "min        0.014354\n",
       "25%        0.033493\n",
       "50%        0.071770\n",
       "75%        0.138756\n",
       "max        0.995215"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(amount).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "executionInfo": {
     "elapsed": 12,
     "status": "ok",
     "timestamp": 1647772017348,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "GigSWsYMeakq",
    "outputId": "a45b5bf6-1f73-44b6-a348-b15f04eb1761"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ENSG00000000003</td>\n",
       "      <td>0.023923</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ENSG00000000419</td>\n",
       "      <td>0.143541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ENSG00000000457</td>\n",
       "      <td>0.344498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ENSG00000000460</td>\n",
       "      <td>0.129187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ENSG00000000938</td>\n",
       "      <td>0.019139</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0         1\n",
       "0  ENSG00000000003  0.023923\n",
       "1  ENSG00000000419  0.143541\n",
       "2  ENSG00000000457  0.344498\n",
       "3  ENSG00000000460  0.129187\n",
       "4  ENSG00000000938  0.019139"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.DataFrame(data=zip(tr_data.index,amount))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "executionInfo": {
     "elapsed": 22259,
     "status": "ok",
     "timestamp": 1647772039596,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "mK4uWs3regyl"
   },
   "outputs": [],
   "source": [
    "assay=[]\n",
    "cnt=[]\n",
    "for i in range(len(df[1])):\n",
    "  if df[1][i] > df[1].quantile(0.25):\n",
    "    assay.append(df[0][i])\n",
    "    cnt.append(df[1][i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 32,
     "status": "ok",
     "timestamp": 1647772039597,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "ZTZ0g4Vrejv-",
    "outputId": "773477d0-a5b1-410b-b241-31714ec2b53f"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10386"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(assay)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 487
    },
    "executionInfo": {
     "elapsed": 27,
     "status": "ok",
     "timestamp": 1647772039597,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "cU8aCBIweqAo",
    "outputId": "923f082a-beaa-41aa-c125-0d2a42059281"
   },
   "outputs": [
    {
     "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>SAMEA5215575</th>\n",
       "      <th>SAMEA5215576</th>\n",
       "      <th>SAMEA5215577</th>\n",
       "      <th>SAMEA5215578</th>\n",
       "      <th>SAMEA5215579</th>\n",
       "      <th>SAMEA5215580</th>\n",
       "      <th>SAMEA5215581</th>\n",
       "      <th>SAMEA5215582</th>\n",
       "      <th>SAMEA5215583</th>\n",
       "      <th>SAMEA5215584</th>\n",
       "      <th>...</th>\n",
       "      <th>SAMEA5215775</th>\n",
       "      <th>SAMEA5215776</th>\n",
       "      <th>SAMEA5215777</th>\n",
       "      <th>SAMEA5215778</th>\n",
       "      <th>SAMEA5215779</th>\n",
       "      <th>SAMEA5215780</th>\n",
       "      <th>SAMEA5215781</th>\n",
       "      <th>SAMEA5215782</th>\n",
       "      <th>SAMEA5215783</th>\n",
       "      <th>SAMEA5215784</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.433432</td>\n",
       "      <td>0.005724</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.19085</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.057303</td>\n",
       "      <td>0.012095</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002232</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002883</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003055</td>\n",
       "      <td>0.002838</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.011174</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.449307</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.009445</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003774</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <td>0.179058</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.068223</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.032667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000286905</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.096603</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000287067</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.055589</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000287977</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.238021</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000288534</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000288579</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10386 rows × 209 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 SAMEA5215575  SAMEA5215576  SAMEA5215577  SAMEA5215578  \\\n",
       "ENSG00000000419      0.000000      0.000000      0.000000      0.000000   \n",
       "ENSG00000000457      0.000000      0.000000      0.002232      0.000000   \n",
       "ENSG00000000460      0.000000      0.000000      0.000000      0.000000   \n",
       "ENSG00000001084      0.000000      0.000000      0.000000      0.000000   \n",
       "ENSG00000001460      0.179058      0.000000      0.068223      0.000000   \n",
       "...                       ...           ...           ...           ...   \n",
       "ENSG00000286905      0.000000      0.096603      1.000000      0.000000   \n",
       "ENSG00000287067      0.000000      0.000000      0.000000      0.000000   \n",
       "ENSG00000287977      0.000000      0.000000      0.000000      0.238021   \n",
       "ENSG00000288534      0.000000      0.000000      0.000000      0.000000   \n",
       "ENSG00000288579      0.000000      0.000000      0.000000      0.000000   \n",
       "\n",
       "                 SAMEA5215579  SAMEA5215580  SAMEA5215581  SAMEA5215582  \\\n",
       "ENSG00000000419      0.000000      0.433432      0.005724           0.0   \n",
       "ENSG00000000457      0.000000      0.002883      0.000000           0.0   \n",
       "ENSG00000000460      0.000000      0.000000      0.011174           0.0   \n",
       "ENSG00000001084      0.000000      0.000000      0.000000           0.0   \n",
       "ENSG00000001460      0.032667      0.000000      0.000000           0.0   \n",
       "...                       ...           ...           ...           ...   \n",
       "ENSG00000286905      0.000000      0.000000      0.000000           0.0   \n",
       "ENSG00000287067      0.000000      0.000000      0.000000           0.0   \n",
       "ENSG00000287977      0.000000      0.000000      0.000000           0.0   \n",
       "ENSG00000288534      0.000000      0.000000      0.000000           0.0   \n",
       "ENSG00000288579      0.000000      0.000000      0.000000           0.0   \n",
       "\n",
       "                 SAMEA5215583  SAMEA5215584  ...  SAMEA5215775  SAMEA5215776  \\\n",
       "ENSG00000000419       0.19085      0.000000  ...      0.057303      0.012095   \n",
       "ENSG00000000457       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000000460       0.00000      0.449307  ...      0.000000      0.009445   \n",
       "ENSG00000001084       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000001460       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "...                       ...           ...  ...           ...           ...   \n",
       "ENSG00000286905       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000287067       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000287977       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000288534       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "ENSG00000288579       0.00000      0.000000  ...      0.000000      0.000000   \n",
       "\n",
       "                 SAMEA5215777  SAMEA5215778  SAMEA5215779  SAMEA5215780  \\\n",
       "ENSG00000000419           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000000457           0.0           0.0      0.003055      0.002838   \n",
       "ENSG00000000460           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000001084           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000001460           0.0           0.0      0.000000      0.000000   \n",
       "...                       ...           ...           ...           ...   \n",
       "ENSG00000286905           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000287067           0.0           0.0      0.000000      0.055589   \n",
       "ENSG00000287977           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000288534           0.0           0.0      0.000000      0.000000   \n",
       "ENSG00000288579           0.0           0.0      0.000000      0.000000   \n",
       "\n",
       "                 SAMEA5215781  SAMEA5215782  SAMEA5215783  SAMEA5215784  \n",
       "ENSG00000000419           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000000457           0.0           0.0      0.000000      0.007048  \n",
       "ENSG00000000460           0.0           0.0      0.003774      0.000000  \n",
       "ENSG00000001084           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000001460           0.0           0.0      0.000000      0.000000  \n",
       "...                       ...           ...           ...           ...  \n",
       "ENSG00000286905           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000287067           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000287977           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000288534           0.0           0.0      0.000000      0.000000  \n",
       "ENSG00000288579           0.0           0.0      0.000000      0.000000  \n",
       "\n",
       "[10386 rows x 209 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "work_data = tr_data.loc[assay]\n",
    "work_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WPN-LSKtimD7"
   },
   "source": [
    "Metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 408
    },
    "executionInfo": {
     "elapsed": 26,
     "status": "ok",
     "timestamp": 1647772039598,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "dO4yD2f9M1pF",
    "outputId": "21af62ee-350d-4776-87cd-36a8fbc547e2"
   },
   "outputs": [
    {
     "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>Assay</th>\n",
       "      <th>Sample Characteristic[organism]</th>\n",
       "      <th>Sample Characteristic[age]</th>\n",
       "      <th>Sample Characteristic[sex]</th>\n",
       "      <th>Sample Characteristic[organism part]</th>\n",
       "      <th>Sample Characteristic[cell type]</th>\n",
       "      <th>Sample Characteristic[phenotype]</th>\n",
       "      <th>Sample Characteristic[individual]</th>\n",
       "      <th>Sample Characteristic[developmental stage]</th>\n",
       "      <th>Factor Value[infect]</th>\n",
       "      <th>Factor Value[time]</th>\n",
       "      <th>Factor Value[phenotype]</th>\n",
       "      <th>Factor Value[single cell identifier]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SAMEA5215575</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51-14-HAP03A09_S27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SAMEA5215576</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51-14-HAP04A04_S17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>SAMEA5215577</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51-14-HAP04B08_S18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SAMEA5215578</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51-14-HAP04C02_S34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SAMEA5215579</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51-14-HAP04C03_S5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Assay Sample Characteristic[organism] Sample Characteristic[age]  \\\n",
       "0  SAMEA5215575                    Homo sapiens                    31 year   \n",
       "1  SAMEA5215576                    Homo sapiens                    31 year   \n",
       "2  SAMEA5215577                    Homo sapiens                    31 year   \n",
       "3  SAMEA5215578                    Homo sapiens                    31 year   \n",
       "4  SAMEA5215579                    Homo sapiens                    31 year   \n",
       "\n",
       "  Sample Characteristic[sex] Sample Characteristic[organism part]  \\\n",
       "0                       male                                blood   \n",
       "1                       male                                blood   \n",
       "2                       male                                blood   \n",
       "3                       male                                blood   \n",
       "4                       male                                blood   \n",
       "\n",
       "  Sample Characteristic[cell type] Sample Characteristic[phenotype]  \\\n",
       "0  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "1  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "2  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "3  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "4  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "\n",
       "  Sample Characteristic[individual]  \\\n",
       "0                              KK51   \n",
       "1                              KK51   \n",
       "2                              KK51   \n",
       "3                              KK51   \n",
       "4                              KK51   \n",
       "\n",
       "  Sample Characteristic[developmental stage] Factor Value[infect]  \\\n",
       "0                                      adult    Influenza B virus   \n",
       "1                                      adult    Influenza B virus   \n",
       "2                                      adult    Influenza B virus   \n",
       "3                                      adult    Influenza B virus   \n",
       "4                                      adult    Influenza B virus   \n",
       "\n",
       "  Factor Value[time]       Factor Value[phenotype]  \\\n",
       "0             14 day  A2/BHA 543 tetramer positive   \n",
       "1             14 day  A2/BHA 543 tetramer positive   \n",
       "2             14 day  A2/BHA 543 tetramer positive   \n",
       "3             14 day  A2/BHA 543 tetramer positive   \n",
       "4             14 day  A2/BHA 543 tetramer positive   \n",
       "\n",
       "  Factor Value[single cell identifier]  \n",
       "0                 KK51-14-HAP03A09_S27  \n",
       "1                 KK51-14-HAP04A04_S17  \n",
       "2                 KK51-14-HAP04B08_S18  \n",
       "3                 KK51-14-HAP04C02_S34  \n",
       "4                  KK51-14-HAP04C03_S5  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp_data = pd.read_csv(dir+'Data/E-MTAB-7606/ExpDesign-E-MTAB-7606.tsv', delimiter=\"\\t\")\n",
    "exp_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 408
    },
    "executionInfo": {
     "elapsed": 26,
     "status": "ok",
     "timestamp": 1647772039599,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "0hPM9IsJM6a0",
    "outputId": "06b34d63-3b5d-4607-ffa2-a90594fd55ca"
   },
   "outputs": [
    {
     "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>Assay</th>\n",
       "      <th>Sample Characteristic[organism]</th>\n",
       "      <th>Sample Characteristic[age]</th>\n",
       "      <th>Sample Characteristic[sex]</th>\n",
       "      <th>Sample Characteristic[organism part]</th>\n",
       "      <th>Sample Characteristic[cell type]</th>\n",
       "      <th>Sample Characteristic[phenotype]</th>\n",
       "      <th>Sample Characteristic[individual]</th>\n",
       "      <th>Sample Characteristic[developmental stage]</th>\n",
       "      <th>Factor Value[infect]</th>\n",
       "      <th>Factor Value[time]</th>\n",
       "      <th>Factor Value[single cell identifier]</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SAMEA5215575</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP03A09_S27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SAMEA5215576</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04A04_S17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>SAMEA5215577</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04B08_S18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SAMEA5215578</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04C02_S34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SAMEA5215579</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04C03_S5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Assay Sample Characteristic[organism] Sample Characteristic[age]  \\\n",
       "0  SAMEA5215575                    Homo sapiens                    31 year   \n",
       "1  SAMEA5215576                    Homo sapiens                    31 year   \n",
       "2  SAMEA5215577                    Homo sapiens                    31 year   \n",
       "3  SAMEA5215578                    Homo sapiens                    31 year   \n",
       "4  SAMEA5215579                    Homo sapiens                    31 year   \n",
       "\n",
       "  Sample Characteristic[sex] Sample Characteristic[organism part]  \\\n",
       "0                       male                                blood   \n",
       "1                       male                                blood   \n",
       "2                       male                                blood   \n",
       "3                       male                                blood   \n",
       "4                       male                                blood   \n",
       "\n",
       "  Sample Characteristic[cell type] Sample Characteristic[phenotype]  \\\n",
       "0  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "1  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "2  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "3  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "4  CD8-positive, alpha-beta T cell     A2/BHA 543 tetramer positive   \n",
       "\n",
       "  Sample Characteristic[individual]  \\\n",
       "0                              KK51   \n",
       "1                              KK51   \n",
       "2                              KK51   \n",
       "3                              KK51   \n",
       "4                              KK51   \n",
       "\n",
       "  Sample Characteristic[developmental stage] Factor Value[infect]  \\\n",
       "0                                      adult    Influenza B virus   \n",
       "1                                      adult    Influenza B virus   \n",
       "2                                      adult    Influenza B virus   \n",
       "3                                      adult    Influenza B virus   \n",
       "4                                      adult    Influenza B virus   \n",
       "\n",
       "  Factor Value[time] Factor Value[single cell identifier]  \n",
       "0             14 day                 KK51-14-HAP03A09_S27  \n",
       "1             14 day                 KK51-14-HAP04A04_S17  \n",
       "2             14 day                 KK51-14-HAP04B08_S18  \n",
       "3             14 day                 KK51-14-HAP04C02_S34  \n",
       "4             14 day                  KK51-14-HAP04C03_S5  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp_data.drop(columns='Factor Value[phenotype]',axis=1,inplace=True)\n",
    "exp_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ft8Vv-z3iqS0"
   },
   "source": [
    "Dropping the assay \"SAMEA5215593\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 26,
     "status": "ok",
     "timestamp": 1647772039600,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "BXl1RmfB0QUR",
    "outputId": "82dc6dac-a91f-40a1-827b-9de8cfd04344"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "209"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp_data.drop(labels=18,axis=0,inplace=True)\n",
    "len(exp_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Jidh61hIjCYU"
   },
   "source": [
    "Renaming column names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 614
    },
    "executionInfo": {
     "elapsed": 22,
     "status": "ok",
     "timestamp": 1647772039600,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "XdQHo_m20S-S",
    "outputId": "569188e8-1b70-4659-d526-e3d9e10b5313"
   },
   "outputs": [
    {
     "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>Assay</th>\n",
       "      <th>Organism</th>\n",
       "      <th>Age</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Organism Part</th>\n",
       "      <th>Cell Type</th>\n",
       "      <th>Phenotype</th>\n",
       "      <th>Individual</th>\n",
       "      <th>Developmental Stage</th>\n",
       "      <th>Infect</th>\n",
       "      <th>Time</th>\n",
       "      <th>Single Cell Identifier</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SAMEA5215575</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP03A09_S27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SAMEA5215576</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04A04_S17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>SAMEA5215577</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04B08_S18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SAMEA5215578</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04C02_S34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SAMEA5215579</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>31 year</td>\n",
       "      <td>male</td>\n",
       "      <td>blood</td>\n",
       "      <td>CD8-positive, alpha-beta T cell</td>\n",
       "      <td>A2/BHA 543 tetramer positive</td>\n",
       "      <td>KK51</td>\n",
       "      <td>adult</td>\n",
       "      <td>Influenza B virus</td>\n",
       "      <td>14 day</td>\n",
       "      <td>KK51-14-HAP04C03_S5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Assay      Organism      Age   Sex Organism Part  \\\n",
       "0  SAMEA5215575  Homo sapiens  31 year  male         blood   \n",
       "1  SAMEA5215576  Homo sapiens  31 year  male         blood   \n",
       "2  SAMEA5215577  Homo sapiens  31 year  male         blood   \n",
       "3  SAMEA5215578  Homo sapiens  31 year  male         blood   \n",
       "4  SAMEA5215579  Homo sapiens  31 year  male         blood   \n",
       "\n",
       "                         Cell Type                     Phenotype Individual  \\\n",
       "0  CD8-positive, alpha-beta T cell  A2/BHA 543 tetramer positive       KK51   \n",
       "1  CD8-positive, alpha-beta T cell  A2/BHA 543 tetramer positive       KK51   \n",
       "2  CD8-positive, alpha-beta T cell  A2/BHA 543 tetramer positive       KK51   \n",
       "3  CD8-positive, alpha-beta T cell  A2/BHA 543 tetramer positive       KK51   \n",
       "4  CD8-positive, alpha-beta T cell  A2/BHA 543 tetramer positive       KK51   \n",
       "\n",
       "  Developmental Stage             Infect    Time Single Cell Identifier  \n",
       "0               adult  Influenza B virus  14 day   KK51-14-HAP03A09_S27  \n",
       "1               adult  Influenza B virus  14 day   KK51-14-HAP04A04_S17  \n",
       "2               adult  Influenza B virus  14 day   KK51-14-HAP04B08_S18  \n",
       "3               adult  Influenza B virus  14 day   KK51-14-HAP04C02_S34  \n",
       "4               adult  Influenza B virus  14 day    KK51-14-HAP04C03_S5  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp_data.columns = ['Assay','Organism','Age','Sex','Organism Part','Cell Type','Phenotype','Individual','Developmental Stage','Infect','Time','Single Cell Identifier']\n",
    "exp_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 22,
     "status": "ok",
     "timestamp": 1647772039601,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "0VG4EUc4LLjW",
    "outputId": "7645c7c9-8f8f-493f-8de2-71b52eaabd26"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.5 year    66\n",
       "14 day      66\n",
       "3 month     15\n",
       "Name: Time, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp_data['Time'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0RsWelBcjOO4"
   },
   "source": [
    "Working data with Genes as rows and Time as columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 316
    },
    "executionInfo": {
     "elapsed": 20,
     "status": "ok",
     "timestamp": 1647772039601,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "J6eP1wYk0gnn",
    "outputId": "573e61e5-2197-41f6-b4c5-f6f2dc4ffe83"
   },
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>...</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "      <th>14 day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.433432</td>\n",
       "      <td>0.005724</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.19085</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.057303</td>\n",
       "      <td>0.012095</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002232</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002883</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003055</td>\n",
       "      <td>0.002838</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.011174</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.449307</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.009445</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003774</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <td>0.179058</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068223</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.032667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 209 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Time               14 day  14 day    14 day  14 day    14 day    14 day  \\\n",
       "ENSG00000000419  0.000000     0.0  0.000000     0.0  0.000000  0.433432   \n",
       "ENSG00000000457  0.000000     0.0  0.002232     0.0  0.000000  0.002883   \n",
       "ENSG00000000460  0.000000     0.0  0.000000     0.0  0.000000  0.000000   \n",
       "ENSG00000001084  0.000000     0.0  0.000000     0.0  0.000000  0.000000   \n",
       "ENSG00000001460  0.179058     0.0  0.068223     0.0  0.032667  0.000000   \n",
       "\n",
       "Time               14 day  14 day   14 day    14 day  ...    14 day    14 day  \\\n",
       "ENSG00000000419  0.005724     0.0  0.19085  0.000000  ...  0.057303  0.012095   \n",
       "ENSG00000000457  0.000000     0.0  0.00000  0.000000  ...  0.000000  0.000000   \n",
       "ENSG00000000460  0.011174     0.0  0.00000  0.449307  ...  0.000000  0.009445   \n",
       "ENSG00000001084  0.000000     0.0  0.00000  0.000000  ...  0.000000  0.000000   \n",
       "ENSG00000001460  0.000000     0.0  0.00000  0.000000  ...  0.000000  0.000000   \n",
       "\n",
       "Time             14 day  14 day    14 day    14 day  14 day  14 day    14 day  \\\n",
       "ENSG00000000419     0.0     0.0  0.000000  0.000000     0.0     0.0  0.000000   \n",
       "ENSG00000000457     0.0     0.0  0.003055  0.002838     0.0     0.0  0.000000   \n",
       "ENSG00000000460     0.0     0.0  0.000000  0.000000     0.0     0.0  0.003774   \n",
       "ENSG00000001084     0.0     0.0  0.000000  0.000000     0.0     0.0  0.000000   \n",
       "ENSG00000001460     0.0     0.0  0.000000  0.000000     0.0     0.0  0.000000   \n",
       "\n",
       "Time               14 day  \n",
       "ENSG00000000419  0.000000  \n",
       "ENSG00000000457  0.007048  \n",
       "ENSG00000000460  0.000000  \n",
       "ENSG00000001084  0.000000  \n",
       "ENSG00000001460  0.000000  \n",
       "\n",
       "[5 rows x 209 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "work_data.columns=exp_data.Time\n",
    "work_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MJluiK4O5u-O"
   },
   "source": [
    "Checking for Trend-Seasonality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 600
    },
    "executionInfo": {
     "elapsed": 1047,
     "status": "ok",
     "timestamp": 1647772040631,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "M4PK9TByAxRk",
    "outputId": "72c8b24d-3cce-49f0-f13c-4ac3ef70ba9b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEICAYAAABPgw/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAYHUlEQVR4nO3df5RcZ33f8fdnZ7VCtmTLtlbGklaSg4VrQUB291gQQiIKJBYlMs0hxEoCJsdF0OK0FNrEOMS4biGUNoU6OClK7WBcjFEbmgiq1KQ2PhCKVctCGCRVSFVldpFtybYWS5asZXe+/WPucmbXszv3zs6P3Wc/r3P27Myd5977feY++5m7z/xSRGBmZrNfV6cLMDOz5nCgm5klwoFuZpYIB7qZWSIc6GZmiXCgm5klwoFu1gSSjkh6U4Prvl7SgWbXZHOPA93aQtJDkk5Iml9gnZB0WSvr6oSJ/YqIb0bE5Z2sydLgQLeWk7QaeD0QwKaOFlOHpO48y8xmIge6tcO7gIeBzwHXjy3Mztr/YdX1d0v62+zyN7LF35V0StKvZ8vfI+mQpGclbZe0rGr9V0j6m+y2pyTdnC2fL+nTko5mP58e+09B0gZJg5J+T9KTwJ/XWpa1faukPZKGJP0vSa+q1VlJV0v6dtbuCUmfkdQzWb/G9le1/hXZfTMkaa+kTVW3fU7SHZL+u6STknZKelljh8VS40C3dngX8IXs55clXVxvhYj4heziqyNiYUR8SdLfA/4QeAdwCfA4cB+ApEXA/wT+B7AMuAx4INvG7wOvAdYBrwauBj5StbuXAhcCq4AttZZJugq4C3gvcBHwWWD7JFNIo8A/A5YArwXeCPzjyfpVvaKkecBXgK8BS4HfAb4gqXpKZjPwL4ELgEPAx2reiTbnONCtpST9PJVQ3BYRjwL/F/iNBjf3m8BdEbE7Is4CHwZem03pvBV4MiL+KCJeiIiTEbGzar3bIuJYRBynEobvrNpuGfhoRJyNiDOTLHsP8NmI2BkRoxFxN3CWygPFOBHxaEQ8HBEjEXGESvj/Ys4+vgZYCHwiIoYj4kHgq1RCfMyXI+J/R8QIlQfJdTm3bYlzoFurXQ98LSKezq7fS9W0S0HLqJyVAxARp4BngOVAH5UHi7rrZZeXVV0/HhEvTFhn4rJVwIeyaZAhSUPZPpdNWA9JL5f0VUlPSnoO+DiVs/U8lgEDEVGeUO/yqutPVl0+TeUBwAw/2WMtI2kBlemRUjYXDTAfWCzp1cDzwDlVq7y0ziaPUgnWse2fS2X640fAAOPPYmuttze7vjJbNqbWR45OXDYAfCwi8kxv/CnwHWBzRJyU9AHg7TnWG6u1T1JXVaivBH6Qc32bw3yGbq30NirzyWupTAusA64AvkllXn0P8KuSzslexnfDhPWfAn6m6vq9wG9LWpfNXX8c2JlNa3wVeKmkD2RPgi6StD5b74vARyT1SloC3AL854J9+TPgfZLWq+JcSX8/m7ufaBHwHHBK0t8B/lGdflXbSeWB7nclzZO0AfgVsucKzKbiQLdWuh7484j4YUQ8OfYDfIbKvPangGEqAXc3lfngarcCd2dTHO+IiAeAPwD+AngCeBlwHUBEnATeTCX8ngQOAm/ItvOvgV3AY8D3gN3ZstwiYheVefTPACeoPBn57kma/3MqzxOcpPJA8KUJt4/r14T9DFN5aedG4GngT4B3RcT/KVKvzU3yF1yYmaXBZ+hmZolwoJuZJcKBbmaWCAe6mVkiOvY69CVLlsTq1as7tXszs1np0UcffToiemvd1rFAX716Nbt27erU7s3MZiVJj092m6dczMwS4UA3M0uEA93MLBEOdDOzRDjQzcwSUTfQJd0l6Zik709yuyTdnn0t2GPZN7u0xGg5eGD/U9z+wEEe2P8Uo2V/Do2Z2Zg8L1v8HJVPmPv8JLdvBNZkP+upfBb0+knaNmy0HLzzzp3sGRjizPAoC3pKrOtbzD03rKfUpWbvzsxs1ql7hh4R3wCenaLJtcDno+JhKl9ecEmzChzz0IFj7BkY4vTwKAGcHh5lz8AQDx041uxdmZnNSs2YQ19O5dtcxgwy/uuyfkrSFkm7JO06fvx4oZ3sPfocZ4ZHxy07MzzKvqPPFSzXzCxNzQj0WvMdNSe3I2JrRPRHRH9vb813rk7qFcvOY0FPadyyBT0l1i47r9B2zMxS1YxAH6TyZbljVjD++xqbYsPlS1nXtxiNDkOUOSebQ99w+dJm78rMbFZqRqBvB96VvdrlNcCPI+KJJmx3nFKXuOeG9fQe/AqLB7/FH2++0k+ImplVqfsqF0lfBDYASyQNAh8F5gFExH8EdgBvofIdi6eB325VsaUucc7QYc4ZOswbr7i4VbsxM5uV6gZ6RGyuc3sA729aRWZm1hC/U9TMLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzROQKdEnXSDog6ZCkm2rcvlLS1yV9R9Jjkt7S/FLNzGwqdQNdUgm4A9gIrAU2S1o7odlHgG0RcSVwHfAnzS7UzMymlucM/WrgUEQcjohh4D7g2gltAjgvu3w+cLR5JZqZWR55An05MFB1fTBbVu1W4LckDQI7gN+ptSFJWyTtkrTr+PHjDZRrZmaTyRPoqrEsJlzfDHwuIlYAbwHukfSibUfE1ojoj4j+3t7e4tWamdmk8gT6INBXdX0FL55SuQHYBhAR3wZeAixpRoFmZpZPnkB/BFgj6VJJPVSe9Nw+oc0PgTcCSLqCSqB7TsXMrI3qBnpEjAA3AvcD+6m8mmWvpNskbcqafQh4j6TvAl8E3h0RE6dlzMyshbrzNIqIHVSe7KxedkvV5X3A65pbmpmZFeF3ipqZJcKBbmaWCAe6mVkiHOhmZolwoJuZJcKBbmaWCAe6mVkiHOhmZolwoJuZJcKBbmaWCAe6mVkiHOhmZolwoJuZJcKBbmaWCAe6mVkiHOhmZolwoJuZJcKBbmaWCAe6mVkiHOhmZolwoJuZJcKBbmaWCAe6mVkiHOhmZolwoJuZJSJXoEu6RtIBSYck3TRJm3dI2idpr6R7m1ummZnV012vgaQScAfwZmAQeETS9ojYV9VmDfBh4HURcULS0lYVbGZmteU5Q78aOBQRhyNiGLgPuHZCm/cAd0TECYCIONbcMs3MrJ48gb4cGKi6Ppgtq/Zy4OWSviXpYUnXNKtAMzPLp+6UC6Aay6LGdtYAG4AVwDclvTIihsZtSNoCbAFYuXJl4WLNzGxyec7QB4G+qusrgKM12vxVRPwkIv4fcIBKwI8TEVsjoj8i+nt7exut2czMasgT6I8AayRdKqkHuA7YPqHNXwJvAJC0hMoUzOFmFmpmZlOrG+gRMQLcCNwP7Ae2RcReSbdJ2pQ1ux94RtI+4OvAv4iIZ1pVtJmZvVieOXQiYgewY8KyW6ouB/DB7MfMzDrA7xQ1M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBG5Al3SNZIOSDok6aYp2r1dUkjqb16JZmaWR91Al1QC7gA2AmuBzZLW1mi3CPgnwM5mF2lmZvXlOUO/GjgUEYcjYhi4D7i2Rrt/BXwSeKGJ9ZmZWU55An05MFB1fTBb9lOSrgT6IuKrU21I0hZJuyTtOn78eOFizcxscnkCXTWWxU9vlLqATwEfqrehiNgaEf0R0d/b25u/SjMzqytPoA8CfVXXVwBHq64vAl4JPCTpCPAaYLufGDUza688gf4IsEbSpZJ6gOuA7WM3RsSPI2JJRKyOiNXAw8CmiNjVkorNzKymuoEeESPAjcD9wH5gW0TslXSbpE2tLtDMzPLpztMoInYAOyYsu2WSthumX5aZmRXld4qamSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlohcgS7pGkkHJB2SdFON2z8oaZ+kxyQ9IGlV80s1M7Op1A10SSXgDmAjsBbYLGnthGbfAfoj4lXAfwU+2exCzcxsannO0K8GDkXE4YgYBu4Drq1uEBFfj4jT2dWHgRXNLdPMzOrpztFmOTBQdX0QWD9F+xuAv651g6QtwBaAlStX5izRbHYbLQcPHTjG3qPP8Ypl57Hh8qWUutTpsixBeQK91siLmg2l3wL6gV+sdXtEbAW2AvT399fchllKRsvBO+/cyZ6BIc4Mj7Kgp8S6vsXcc8N6h7o1XZ4pl0Ggr+r6CuDoxEaS3gT8PrApIs42pzyz2e2hA8fYMzDE6eFRAjg9PMqegSEeOnCs06VZgvIE+iPAGkmXSuoBrgO2VzeQdCXwWSph7pFqltl79DnODI+OW3ZmeJR9R5/rUEWWsrqBHhEjwI3A/cB+YFtE7JV0m6RNWbN/CywE/oukPZK2T7I5sznlFcvOY0FPadyyBT0l1i47r0MVWcryzKETETuAHROW3VJ1+U1NrsssCRsuX8q6vsV8+wdPEF3dnDN/Huv6FrPh8qWdLs0S5HeKmrVQqUvcc8N6eg9+hcWD3+KPN1/pJ0StZXKdoZtZ40pd4pyhw5wzdJg3XnFxp8uxhPkM3cwsET5DN5vBIqb/do1GNlFklWbUmHtfbdtTY/fbuPUnqbYk0V1qzbl0UoFeLgejTRxczdrUZAe2Wdr49zSlqeqodx80uw/N2FwjQTXZGqPlyj3w7PPDjJTLjJaDkdGo/C6P/S6Puz5ajhlzbK15lizsYc3Fi1qy7aQC/UdDZxg8cabTZZi9yPNnRwA48OTJDldiKfMcuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mlggHuplZIhzoZmaJcKCbmSXCgW5mloikvuDCLBXlcrBnYIgjzzzP6ovOZV3fYrq61OmybIZzoJvNMOVy8PG/3s+hY6cYHinT093FZUsXcvPGKxzqNiVPuXRAuRzsfvwEX949yO7HT1Au+4sjJzMX76s9A0McOnaKsyNlAjg7UubQsVPsGRjqdGk2w+U6Q5d0DfAfgBLwnyLiExNunw98Hvi7wDPAr0fEkeaWmgaffeU3V++rI888z/BIedyy4ZEyR555nqtWXdChqmw2qHuGLqkE3AFsBNYCmyWtndDsBuBERFwGfAr4N80uNBU++8pvrt5Xqy86l57u8X+aPd1drL7o3A5VZLOFIqb+F1bSa4FbI+KXs+sfBoiIP6xqc3/W5tuSuoEngd6YYuMXrroi3nzzXYUL3vPdPQCse/W6F912dqTM2ZHRwttsp+Mnz/L0qeEXLe9d2MOSRfM7UFFFRHDq7Cgv/GSUl8wrsXB+CamzZ8Ez9b5qxMF93wdgzdpX1m0bEfzw2TOcPvsTQKhLLJhXYuWFCzp+TIpox5iaieO2nnmlLhbMKzW8/rb3/dyjEdFf67Y8Uy7LgYGq64PA+snaRMSIpB8DFwFPVzeStAXYArDwkpflKn6iWkFeT9E/ph8cOgylHpYtuyT3AMm7j5fMKyFB9UOdBPPrHOAifSi6znQCpGhdRdq3675q5X07psi2JbHywgWcOtvD2Z+MMj9nUM2kfrdjTM3UcdtI+2bJE+i17pmJZ9552hARW4GtAP39/fGl9742x+7zG3j2NIMnzoxbVi4H773/TkYXXsxbf/aXpnz519icbfkli6HUzfGTZzl/Qb452/ff+2EAbvnd7VO2a3ReOO/2q/eTt9+7Hz/B7Q8eBFX+zY+A0XLwK69aXnfOtmhdRdo3cl8V6Xcj7RvpRzs00o9G+pB3nUbHVDvGbTvuq6naL1nYw5qLF+XaTi3b3jf5bXkCfRDoq7q+Ajg6SZvBbMrlfODZQlW2wFggnFr7Nih1c/uDB6cMhLE5W7p7gPFztvUGyPBFlzG68GJ2P35iygHS1SVu3nhFodcYF9l+I/2eqU/CFb2viva7aPuZqpF+FB1TRddpZEy1Y9y245g3ct82S56XLT4CrJF0qaQe4Dpg4sPOduD67PLbgQenmj9vl3EBra66T6pNNUAmUz1Azlz6em5/8GDlLH+Kl9d1dYmrVl3Ar161gqtWXZA7nPNuv2i/G30Sbmzgnln1ulwvKSzaHordV0X7XbT9dPrRSkX70ciYKrpOI2OqHeO2kWNe5Hg3ct82U91Aj4gR4EbgfmA/sC0i9kq6TdKmrNmdwEWSDgEfBG5qVcFFFA3odg2QIhrZftF+r+tbzGVLFzK/uwsB87OpjXV9iyfdR9GB246BXrTf7XoAb7Wi/WhkTBVdp5Ex1Y5xW3QfRY93q/OgnlyvQ4+IHcCOCctuqbr8AvBrzS1t+sYC+mzVAZwqoMcGyMQ520YHSDOmKxrZftF+NzINVHR6qtHprCKK9rto+3b1o6ii/WhkTBVdp5Ex1Y5xW3QfRY93p6cvk37rf9GAbscAKaqR7TfywDQ2tZF30BUduO0Y6EX7PRMfwBtRtB+NjKlG1ik6ptoxbovuo+jxbnUe1JN0oDcS0O0YhEU0OsiL9ruodpwNF1W03zPxAbwRRfvRyJhq9ThvpB/t2Ec7/stvprpvLGqV/v7+2LVrV1O3Wetli+3Q6k/Gm4mfvFf0JYWpvI0/pX4UHVMzcRy2WqMvnZ3qfpruyxYlTfrGIge6NazoH3gqgZBKPyyfZh9vB3pODnQzm+laGej++Fwzs0Q40M3MEuFANzNLhAPdzCwRDnQzs0Q40M3MEuFANzNLhAPdzCwRDnQzs0Q40M3MEuFANzNLhAPdzCwRSX0eendJL/oKudZo/APNWvlZaI1uuhU1Tfahb/V21flvojWbvZIK9EvOX8Al5y/odBk2wzXrE0bzbKa6SUQwUg5Gy9W/y5Xfo7WXt+TBtkjbggV08vG4PScDje2kurZSCz9qOalAN8tDas4fVPHNiO5SU3ZtVpPn0M3MEuFANzNLhAPdzCwRDnQzs0Q40M3MEuFANzNLhAPdzCwRDnQzs0Q40M3MEqFmvQ268I6l48DjDa6+BHi6ieXMFnO13zB3++5+zy15+r0qInpr3dCxQJ8OSbsior/TdbTbXO03zN2+u99zy3T77SkXM7NEONDNzBIxWwN9a6cL6JC52m+Yu313v+eWafV7Vs6hm5nZi83WM3QzM5vAgW5mlohZF+iSrpF0QNIhSTd1up52kXRE0vck7ZG0q9P1tIqkuyQdk/T9qmUXSvobSQez3xd0ssZWmKTft0r6UXbM90h6SydrbAVJfZK+Lmm/pL2S/mm2POljPkW/p3XMZ9UcuqQS8APgzcAg8AiwOSL2dbSwNpB0BOiPiKTfbCHpF4BTwOcj4pXZsk8Cz0bEJ7IH8Qsi4vc6WWezTdLvW4FTEfHvOllbK0m6BLgkInZLWgQ8CrwNeDcJH/Mp+v0OpnHMZ9sZ+tXAoYg4HBHDwH3AtR2uyZooIr4BPDth8bXA3dnlu6kM/KRM0u/kRcQTEbE7u3wS2A8sJ/FjPkW/p2W2BfpyYKDq+iBNuBNmiQC+JulRSVs6XUybXRwRT0DlDwFY2uF62ulGSY9lUzJJTTtMJGk1cCWwkzl0zCf0G6ZxzGdboNf6nvXZM2c0Pa+LiKuAjcD7s3/RLW1/CrwMWAc8AfxRZ8tpHUkLgb8APhARz3W6nnap0e9pHfPZFuiDQF/V9RXA0Q7V0lYRcTT7fQz4b1Smn+aKp7I5x7G5x2MdrqctIuKpiBiNiDLwZyR6zCXNoxJqX4iIL2eLkz/mtfo93WM+2wL9EWCNpEsl9QDXAds7XFPLSTo3e+IESecCvwR8f+q1krIduD67fD3wVx2spW3GAi3zD0jwmEsScCewPyL+fdVNSR/zyfo93WM+q17lApC9jOfTQAm4KyI+1uGSWk7Sz1A5KwfoBu5Ntd+SvghsoPIxok8BHwX+EtgGrAR+CPxaRCT1BOIk/d5A5V/vAI4A7x2bV06FpJ8Hvgl8Dyhni2+mMp+c7DGfot+bmcYxn3WBbmZmtc22KRczM5uEA93MLBEOdDOzRDjQzcwS4UA3M0uEA93MLBEOdDOzRPx/WzaGF9BemLcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#%%script false\n",
    "from matplotlib import pyplot\n",
    "from statsmodels.graphics.tsaplots import plot_acf,plot_pacf\n",
    "plot_acf(work_data.iloc[100])\n",
    "plot_pacf(work_data.iloc[100])\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1JsJKHli4LXl"
   },
   "source": [
    "Extracting control data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 300
    },
    "executionInfo": {
     "elapsed": 39,
     "status": "ok",
     "timestamp": 1647772040632,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "mJGEkG4U1ITj",
    "outputId": "2edae1f9-ffa9-4682-d6f0-b4b8ae80a460"
   },
   "outputs": [
    {
     "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>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>...</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002642</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.421985</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.034608</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.005155</td>\n",
       "      <td>0.004443</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002648</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001721</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002998</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.01764</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.073995</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.03484</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.044574</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 62 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Control   Control   Control  Control  Control   Control  \\\n",
       "ENSG00000000419      0.0  0.002642  0.000000  0.00000      0.0  0.000000   \n",
       "ENSG00000000457      0.0  0.005155  0.004443  0.00000      0.0  0.000000   \n",
       "ENSG00000000460      0.0  0.000000  0.000000  0.01764      0.0  0.073995   \n",
       "ENSG00000001084      0.0  0.000000  0.000000  0.00000      0.0  0.000000   \n",
       "ENSG00000001460      0.0  0.000000  0.000000  0.00000      0.0  0.000000   \n",
       "\n",
       "                 Control  Control   Control  Control  ...  Control   Control  \\\n",
       "ENSG00000000419  0.00000      0.0  0.000000      0.0  ...      0.0  0.421985   \n",
       "ENSG00000000457  0.00000      0.0  0.002648      0.0  ...      0.0  0.000000   \n",
       "ENSG00000000460  0.00000      0.0  0.000000      0.0  ...      0.0  0.000000   \n",
       "ENSG00000001084  0.00000      0.0  0.000000      0.0  ...      0.0  0.000000   \n",
       "ENSG00000001460  0.03484      0.0  0.000000      0.0  ...      0.0  0.044574   \n",
       "\n",
       "                 Control  Control   Control   Control  Control  Control  \\\n",
       "ENSG00000000419      0.0      0.0  0.000000  0.034608      0.0      0.0   \n",
       "ENSG00000000457      0.0      0.0  0.001721  0.000000      0.0      0.0   \n",
       "ENSG00000000460      0.0      0.0  0.000000  0.000000      0.0      0.0   \n",
       "ENSG00000001084      0.0      0.0  0.000000  0.000000      0.0      0.0   \n",
       "ENSG00000001460      0.0      0.0  0.000000  0.000000      0.0      0.0   \n",
       "\n",
       "                  Control  Control  \n",
       "ENSG00000000419  0.000000      0.0  \n",
       "ENSG00000000457  0.002998      0.0  \n",
       "ENSG00000000460  0.000000      0.0  \n",
       "ENSG00000001084  0.000000      0.0  \n",
       "ENSG00000001460  0.000000      0.0  \n",
       "\n",
       "[5 rows x 62 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "control_data = work_data.drop(['14 day','1.5 year','3 month'],axis=1)\n",
    "control_data.columns=['Control' for i in range(len(control_data.columns))]\n",
    "control_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "executionInfo": {
     "elapsed": 34,
     "status": "ok",
     "timestamp": 1647772040632,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "9YRIkrvy4WiJ"
   },
   "outputs": [],
   "source": [
    "X_data = pd.concat([control_data,work_data[['14 day','3 month']]],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 316
    },
    "executionInfo": {
     "elapsed": 33,
     "status": "ok",
     "timestamp": 1647772040633,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "7AGJ7mH94zeT",
    "outputId": "b83d5752-d497-4489-d53d-a6462aa551c2"
   },
   "outputs": [
    {
     "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>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>Control</th>\n",
       "      <th>...</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "      <th>3 month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002642</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.005155</td>\n",
       "      <td>0.004443</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002648</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001855</td>\n",
       "      <td>0.005152</td>\n",
       "      <td>0.003880</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00456</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.01764</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.073995</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.005315</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.03484</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.038445</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.27396</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.549253</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 143 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Control   Control   Control  Control  Control   Control  \\\n",
       "ENSG00000000419      0.0  0.002642  0.000000  0.00000      0.0  0.000000   \n",
       "ENSG00000000457      0.0  0.005155  0.004443  0.00000      0.0  0.000000   \n",
       "ENSG00000000460      0.0  0.000000  0.000000  0.01764      0.0  0.073995   \n",
       "ENSG00000001084      0.0  0.000000  0.000000  0.00000      0.0  0.000000   \n",
       "ENSG00000001460      0.0  0.000000  0.000000  0.00000      0.0  0.000000   \n",
       "\n",
       "                 Control  Control   Control  Control  ...   3 month   3 month  \\\n",
       "ENSG00000000419  0.00000      0.0  0.000000      0.0  ...  0.000000  0.000000   \n",
       "ENSG00000000457  0.00000      0.0  0.002648      0.0  ...  0.001855  0.005152   \n",
       "ENSG00000000460  0.00000      0.0  0.000000      0.0  ...  0.000000  0.000000   \n",
       "ENSG00000001084  0.00000      0.0  0.000000      0.0  ...  0.000000  0.000000   \n",
       "ENSG00000001460  0.03484      0.0  0.000000      0.0  ...  0.000000  0.000000   \n",
       "\n",
       "                  3 month   3 month  3 month  3 month  3 month  3 month  \\\n",
       "ENSG00000000419  0.000000  0.000000  0.00000      0.0  0.00000      0.0   \n",
       "ENSG00000000457  0.003880  0.000000  0.00456      0.0  0.00000      0.0   \n",
       "ENSG00000000460  0.005315  0.000000  0.00000      0.0  0.00000      0.0   \n",
       "ENSG00000001084  0.000000  0.000000  0.00000      0.0  0.00000      0.0   \n",
       "ENSG00000001460  0.000000  0.038445  0.00000      0.0  0.27396      0.0   \n",
       "\n",
       "                  3 month  3 month  \n",
       "ENSG00000000419  0.000000      0.0  \n",
       "ENSG00000000457  0.000000      0.0  \n",
       "ENSG00000000460  0.000000      0.0  \n",
       "ENSG00000001084  0.000000      0.0  \n",
       "ENSG00000001460  0.549253      0.0  \n",
       "\n",
       "[5 rows x 143 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 316
    },
    "executionInfo": {
     "elapsed": 30,
     "status": "ok",
     "timestamp": 1647772040634,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "oD_-SGXN460i",
    "outputId": "d54dc926-6897-42ca-9494-8a4be2fd0dd6"
   },
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>...</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "      <th>1.5 year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ENSG00000000419</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.653975</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.612043</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000457</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003356</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002662</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0021</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002882</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.003906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000000460</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.042184</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001084</th>\n",
       "      <td>0.626635</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ENSG00000001460</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.02229</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.034453</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.017486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 66 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Time             1.5 year  1.5 year  1.5 year  1.5 year  1.5 year  1.5 year  \\\n",
       "ENSG00000000419  0.000000       0.0       0.0  0.000000       0.0  0.000000   \n",
       "ENSG00000000457  0.000000       0.0       0.0  0.003356       0.0  0.002662   \n",
       "ENSG00000000460  0.000000       0.0       0.0  0.000000       0.0  0.000000   \n",
       "ENSG00000001084  0.626635       0.0       0.0  0.000000       0.0  0.000000   \n",
       "ENSG00000001460  0.000000       0.0       0.0  0.000000       0.0  0.000000   \n",
       "\n",
       "Time             1.5 year  1.5 year  1.5 year  1.5 year  ...  1.5 year  \\\n",
       "ENSG00000000419       0.0  0.653975       0.0       0.0  ...  0.612043   \n",
       "ENSG00000000457       0.0  0.000000       0.0       0.0  ...  0.000000   \n",
       "ENSG00000000460       0.0  0.042184       1.0       0.0  ...  0.000000   \n",
       "ENSG00000001084       0.0  0.000000       0.0       0.0  ...  0.000000   \n",
       "ENSG00000001460       0.0  0.000000       0.0       0.0  ...  0.000000   \n",
       "\n",
       "Time             1.5 year  1.5 year  1.5 year  1.5 year  1.5 year  1.5 year  \\\n",
       "ENSG00000000419    0.0000       0.0   0.00000       0.0  0.000000       0.0   \n",
       "ENSG00000000457    0.0021       0.0   0.00000       0.0  0.002882       0.0   \n",
       "ENSG00000000460    0.0000       0.0   0.00000       0.0  0.000000       0.0   \n",
       "ENSG00000001084    0.0000       0.0   0.00000       0.0  0.000000       0.0   \n",
       "ENSG00000001460    0.0000       0.0   0.02229       0.0  0.034453       0.0   \n",
       "\n",
       "Time             1.5 year  1.5 year  1.5 year  \n",
       "ENSG00000000419       0.0       0.0  0.000000  \n",
       "ENSG00000000457       0.0       0.0  0.003906  \n",
       "ENSG00000000460       0.0       0.0  0.000000  \n",
       "ENSG00000001084       0.0       0.0  0.000000  \n",
       "ENSG00000001460       0.0       0.0  0.017486  \n",
       "\n",
       "[5 rows x 66 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_data = work_data['1.5 year']\n",
    "y_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 28,
     "status": "ok",
     "timestamp": 1647772040635,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "GsWqc-GH5JUa",
    "outputId": "7f0d8933-ad4d-450c-93c2-d94611fa7a5e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(10386, 143)\n",
      "(10386, 66)\n"
     ]
    }
   ],
   "source": [
    "print(np.shape(X_data))\n",
    "print(np.shape(y_data))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DSR0JRBt5u-W"
   },
   "source": [
    "# DNN regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "executionInfo": {
     "elapsed": 434,
     "status": "ok",
     "timestamp": 1647772054440,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "DrpdM2-01TWR"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "import keras\n",
    "from keras.utils.vis_utils import plot_model\n",
    "from keras import regularizers\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense,Dropout\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import r2_score\n",
    "import tensorflow as tf\n",
    "from keras.models import Model\n",
    "from keras.layers import Input, Dense\n",
    "from scipy.stats import pearsonr, spearmanr\n",
    "from sklearn.metrics.pairwise import euclidean_distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.keras.backend.set_floatx('float64')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DQUYbYMI5u-X"
   },
   "source": [
    "**Finding the best seed**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "executionInfo": {
     "elapsed": 4,
     "status": "ok",
     "timestamp": 1647772055792,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "OLmCpvxb5u-Y"
   },
   "outputs": [],
   "source": [
    "def MLP1(x_shape,y_shape):\n",
    "    model = Sequential()\n",
    "    model.add(Dense(130, input_dim=x_shape, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dense(75, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dense(y_shape, kernel_initializer='uniform', activation=\"linear\"))\n",
    "    model.compile(optimizer='adam',loss='mean_squared_error', metrics=['mse'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 231226,
     "status": "ok",
     "timestamp": 1647772288818,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "xVcu5lby5u-Y",
    "outputId": "55373583-dbd9-4516-80fd-75b4b16d8ec2",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Seed = 86\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 16\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 65\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0128 - val_mse: 0.0128\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0128 - val_mse: 0.0128\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0128 - val_mse: 0.0128\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0128 - val_mse: 0.0128\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0128 - val_mse: 0.0128\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "\n",
      "Seed = 35\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Seed = 15\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 88\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "\n",
      "Seed = 23\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 58\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 91\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0127 - val_mse: 0.0127\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0126 - val_mse: 0.0126\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0127 - val_mse: 0.0127\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0127 - val_mse: 0.0127\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0127 - val_mse: 0.0127\n",
      "\n",
      "Seed = 56\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "\n",
      "Seed = 25\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "\n",
      "Seed = 13\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0124 - mse: 0.0124 - val_loss: 0.0136 - val_mse: 0.0136\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0124 - mse: 0.0124 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "\n",
      "Seed = 67\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0142 - val_mse: 0.0142\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0141 - val_mse: 0.0141\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0140 - val_mse: 0.0140\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "\n",
      "Seed = 44\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "\n",
      "Seed = 12\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0137 - val_mse: 0.0137\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0138 - val_mse: 0.0138\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0140 - val_mse: 0.0140\n",
      "\n",
      "Seed = 81\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0141 - val_mse: 0.0141\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0140 - val_mse: 0.0140\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0141 - val_mse: 0.0141\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0139 - val_mse: 0.0139\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0124 - mse: 0.0124 - val_loss: 0.0140 - val_mse: 0.0140\n",
      "\n",
      "Seed = 34\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Seed = 90\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0135 - val_mse: 0.0135\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "\n",
      "Seed = 11\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Seed = 32\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n"
     ]
    }
   ],
   "source": [
    "rnd_state,r2scores,msescores,maescores,pearsoncorr = [],[],[],[],[]\n",
    "for i in np.random.randint(1,101,20):\n",
    "    print('\\nSeed = {}'.format(i))\n",
    "    X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=i)\n",
    "    model = MLP1(np.shape(X_train)[1],np.shape(y_train)[1])\n",
    "    model.fit(np.array(X_train), np.array(y_train), epochs=10, batch_size=16, validation_split=0.2, shuffle=True)\n",
    "    y_pred = model.predict(np.array(X_test))\n",
    "    \n",
    "    #results\n",
    "    rnd_state.append(i)\n",
    "    \n",
    "    actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    r2scores.append(r2_score(actual_mean, pred_mean))\n",
    "    msescores.append(mean_squared_error(np.array(y_test), y_pred))\n",
    "    maescores.append(mean_absolute_error(np.array(y_test), y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 707
    },
    "executionInfo": {
     "elapsed": 41,
     "status": "ok",
     "timestamp": 1647772288819,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "2ePntufW5u-i",
    "outputId": "6fe93048-f081-4d56-9095-f3353f81ccac"
   },
   "outputs": [
    {
     "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>Seed</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>86.0</td>\n",
       "      <td>0.531851</td>\n",
       "      <td>0.013242</td>\n",
       "      <td>0.045483</td>\n",
       "      <td>0.532838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>16.0</td>\n",
       "      <td>0.654590</td>\n",
       "      <td>0.013372</td>\n",
       "      <td>0.043334</td>\n",
       "      <td>0.676038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>65.0</td>\n",
       "      <td>0.544518</td>\n",
       "      <td>0.013446</td>\n",
       "      <td>0.046218</td>\n",
       "      <td>0.548951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.0</td>\n",
       "      <td>0.588611</td>\n",
       "      <td>0.013478</td>\n",
       "      <td>0.045293</td>\n",
       "      <td>0.618095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15.0</td>\n",
       "      <td>0.534632</td>\n",
       "      <td>0.013416</td>\n",
       "      <td>0.046059</td>\n",
       "      <td>0.596124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>88.0</td>\n",
       "      <td>0.545778</td>\n",
       "      <td>0.012956</td>\n",
       "      <td>0.044691</td>\n",
       "      <td>0.609377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>23.0</td>\n",
       "      <td>0.511565</td>\n",
       "      <td>0.013538</td>\n",
       "      <td>0.048859</td>\n",
       "      <td>0.602038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>58.0</td>\n",
       "      <td>0.434182</td>\n",
       "      <td>0.013483</td>\n",
       "      <td>0.045795</td>\n",
       "      <td>0.525353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>91.0</td>\n",
       "      <td>0.571390</td>\n",
       "      <td>0.013413</td>\n",
       "      <td>0.045497</td>\n",
       "      <td>0.602705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>56.0</td>\n",
       "      <td>0.645687</td>\n",
       "      <td>0.013031</td>\n",
       "      <td>0.045276</td>\n",
       "      <td>0.637324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>25.0</td>\n",
       "      <td>0.615785</td>\n",
       "      <td>0.013114</td>\n",
       "      <td>0.042564</td>\n",
       "      <td>0.622964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>13.0</td>\n",
       "      <td>0.589015</td>\n",
       "      <td>0.013392</td>\n",
       "      <td>0.043595</td>\n",
       "      <td>0.600303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>67.0</td>\n",
       "      <td>0.662407</td>\n",
       "      <td>0.012623</td>\n",
       "      <td>0.044350</td>\n",
       "      <td>0.634235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>44.0</td>\n",
       "      <td>0.601546</td>\n",
       "      <td>0.013149</td>\n",
       "      <td>0.046302</td>\n",
       "      <td>0.595128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>12.0</td>\n",
       "      <td>0.519943</td>\n",
       "      <td>0.012598</td>\n",
       "      <td>0.048132</td>\n",
       "      <td>0.601602</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>81.0</td>\n",
       "      <td>0.542082</td>\n",
       "      <td>0.013165</td>\n",
       "      <td>0.045916</td>\n",
       "      <td>0.580693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>34.0</td>\n",
       "      <td>0.463736</td>\n",
       "      <td>0.013146</td>\n",
       "      <td>0.045294</td>\n",
       "      <td>0.473511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>90.0</td>\n",
       "      <td>0.556129</td>\n",
       "      <td>0.013227</td>\n",
       "      <td>0.043636</td>\n",
       "      <td>0.574576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>11.0</td>\n",
       "      <td>0.552307</td>\n",
       "      <td>0.013170</td>\n",
       "      <td>0.042990</td>\n",
       "      <td>0.534188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>32.0</td>\n",
       "      <td>0.545351</td>\n",
       "      <td>0.013264</td>\n",
       "      <td>0.045765</td>\n",
       "      <td>0.553919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>47.1</td>\n",
       "      <td>0.560555</td>\n",
       "      <td>0.013211</td>\n",
       "      <td>0.045252</td>\n",
       "      <td>0.585998</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Seed   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0    86.0  0.531851  0.013242  0.045483            0.532838\n",
       "1    16.0  0.654590  0.013372  0.043334            0.676038\n",
       "2    65.0  0.544518  0.013446  0.046218            0.548951\n",
       "3    35.0  0.588611  0.013478  0.045293            0.618095\n",
       "4    15.0  0.534632  0.013416  0.046059            0.596124\n",
       "5    88.0  0.545778  0.012956  0.044691            0.609377\n",
       "6    23.0  0.511565  0.013538  0.048859            0.602038\n",
       "7    58.0  0.434182  0.013483  0.045795            0.525353\n",
       "8    91.0  0.571390  0.013413  0.045497            0.602705\n",
       "9    56.0  0.645687  0.013031  0.045276            0.637324\n",
       "10   25.0  0.615785  0.013114  0.042564            0.622964\n",
       "11   13.0  0.589015  0.013392  0.043595            0.600303\n",
       "12   67.0  0.662407  0.012623  0.044350            0.634235\n",
       "13   44.0  0.601546  0.013149  0.046302            0.595128\n",
       "14   12.0  0.519943  0.012598  0.048132            0.601602\n",
       "15   81.0  0.542082  0.013165  0.045916            0.580693\n",
       "16   34.0  0.463736  0.013146  0.045294            0.473511\n",
       "17   90.0  0.556129  0.013227  0.043636            0.574576\n",
       "18   11.0  0.552307  0.013170  0.042990            0.534188\n",
       "19   32.0  0.545351  0.013264  0.045765            0.553919\n",
       "avg  47.1  0.560555  0.013211  0.045252            0.585998"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(rnd_state,r2scores,msescores,maescores,pearsoncorr))\n",
    "result.columns = ['Seed','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(rnd_state)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(rnd_state):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "executionInfo": {
     "elapsed": 410,
     "status": "ok",
     "timestamp": 1647772322177,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "cl-zWe0P1Mbh"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/DNN_bestSeed.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WzbFXRUvqXqH"
   },
   "source": [
    "Extracting results with repeated runs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "w3UwkFVs_-dS"
   },
   "source": [
    "**DNN with no dropout**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 157582,
     "status": "ok",
     "timestamp": 1647772497678,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "ZdqElqDrqLTH",
    "outputId": "f9b56286-db98-4214-c440-3a94d8581e27"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Run = 1\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 3s 5ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 2\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 3\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 4\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 3s 5ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 2s 6ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 5\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 6\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 2s 4ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 2s 5ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 7\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 8\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 9\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 10\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0127 - mse: 0.0127 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0126 - mse: 0.0126 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0125 - mse: 0.0125 - val_loss: 0.0130 - val_mse: 0.0130\n"
     ]
    }
   ],
   "source": [
    "run,r2scores,msescores,maescores,pearsoncorr = [],[],[],[],[]\n",
    "for i in range(10):\n",
    "    print('\\nRun = {}'.format(i+1))\n",
    "    X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "    model = MLP1(np.shape(X_train)[1],np.shape(y_train)[1])\n",
    "    model.fit(np.array(X_train), np.array(y_train), epochs=10, batch_size=16, validation_split=0.2, shuffle=True)\n",
    "    y_pred = model.predict(np.array(X_test))\n",
    "    \n",
    "    #results\n",
    "    run.append(i+1)\n",
    "    \n",
    "    actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    r2scores.append(r2_score(actual_mean, pred_mean))\n",
    "    msescores.append(mean_squared_error(np.array(y_test), y_pred))\n",
    "    maescores.append(mean_absolute_error(np.array(y_test), y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 57,
     "status": "ok",
     "timestamp": 1647772497679,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "ygFLW9CyqMyx",
    "outputId": "c76607d0-c3b1-46bf-e607-68bc956b97d4"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-9620eef2-b403-4031-b02e-7fd085fe5b07\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>#Run</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.658248</td>\n",
       "      <td>0.013355</td>\n",
       "      <td>0.045400</td>\n",
       "      <td>0.699071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.654771</td>\n",
       "      <td>0.013384</td>\n",
       "      <td>0.043674</td>\n",
       "      <td>0.674721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.643683</td>\n",
       "      <td>0.013428</td>\n",
       "      <td>0.045500</td>\n",
       "      <td>0.692920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "      <td>0.639410</td>\n",
       "      <td>0.013435</td>\n",
       "      <td>0.046342</td>\n",
       "      <td>0.684709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>0.671129</td>\n",
       "      <td>0.013331</td>\n",
       "      <td>0.046584</td>\n",
       "      <td>0.701345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.664399</td>\n",
       "      <td>0.013354</td>\n",
       "      <td>0.044198</td>\n",
       "      <td>0.692056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7.0</td>\n",
       "      <td>0.678102</td>\n",
       "      <td>0.013331</td>\n",
       "      <td>0.044698</td>\n",
       "      <td>0.695824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8.0</td>\n",
       "      <td>0.647746</td>\n",
       "      <td>0.013380</td>\n",
       "      <td>0.043714</td>\n",
       "      <td>0.680988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9.0</td>\n",
       "      <td>0.639543</td>\n",
       "      <td>0.013417</td>\n",
       "      <td>0.047182</td>\n",
       "      <td>0.688590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10.0</td>\n",
       "      <td>0.685486</td>\n",
       "      <td>0.013293</td>\n",
       "      <td>0.046760</td>\n",
       "      <td>0.693668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>5.5</td>\n",
       "      <td>0.658252</td>\n",
       "      <td>0.013371</td>\n",
       "      <td>0.045405</td>\n",
       "      <td>0.690389</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9620eef2-b403-4031-b02e-7fd085fe5b07')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-9620eef2-b403-4031-b02e-7fd085fe5b07 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-9620eef2-b403-4031-b02e-7fd085fe5b07');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "     #Run   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     1.0  0.658248  0.013355  0.045400            0.699071\n",
       "1     2.0  0.654771  0.013384  0.043674            0.674721\n",
       "2     3.0  0.643683  0.013428  0.045500            0.692920\n",
       "3     4.0  0.639410  0.013435  0.046342            0.684709\n",
       "4     5.0  0.671129  0.013331  0.046584            0.701345\n",
       "5     6.0  0.664399  0.013354  0.044198            0.692056\n",
       "6     7.0  0.678102  0.013331  0.044698            0.695824\n",
       "7     8.0  0.647746  0.013380  0.043714            0.680988\n",
       "8     9.0  0.639543  0.013417  0.047182            0.688590\n",
       "9    10.0  0.685486  0.013293  0.046760            0.693668\n",
       "avg   5.5  0.658252  0.013371  0.045405            0.690389"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(run,r2scores,msescores,maescores,pearsoncorr))\n",
    "result.columns = ['#Run','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(run)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(run):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 682
    },
    "executionInfo": {
     "elapsed": 20,
     "status": "ok",
     "timestamp": 1647772497680,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "bVng80r_o_qE",
    "outputId": "7b48af43-3482-42a3-bf77-a1abf3851f2a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_29\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense_87 (Dense)            (None, 130)               18720     \n",
      "                                                                 \n",
      " dense_88 (Dense)            (None, 75)                9825      \n",
      "                                                                 \n",
      " dense_89 (Dense)            (None, 66)                5016      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 33,561\n",
      "Trainable params: 33,561\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.summary()\n",
    "plot_model(model, show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "executionInfo": {
     "elapsed": 12,
     "status": "ok",
     "timestamp": 1647772497680,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "4BpzVYFv5u-j"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/DNN_noDropout_10runs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dYZgE-OO5u-n"
   },
   "source": [
    "**DNN with 10% dropout all all layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "executionInfo": {
     "elapsed": 12,
     "status": "ok",
     "timestamp": 1647772497681,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "LxwYFii15u-n"
   },
   "outputs": [],
   "source": [
    "def MLP2(x_shape,y_shape,rnd):\n",
    "    model = Sequential()\n",
    "    model.add(Dropout(0.1,seed=rnd))\n",
    "    model.add(Dense(130, input_dim=x_shape, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dropout(0.1,seed=rnd))\n",
    "    model.add(Dense(75, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dropout(0.1,seed=rnd))\n",
    "    model.add(Dense(y_shape, kernel_initializer='uniform', activation=\"linear\"))\n",
    "    model.compile(optimizer='adam',loss='mean_squared_error', metrics=['mse'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 187515,
     "status": "ok",
     "timestamp": 1647772685184,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "8KRmu1SO5u-n",
    "outputId": "e5683a0b-c0f8-4b07-fa4b-c5e88c91eaf5",
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Run = 1\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "\n",
      "Run = 2\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 3\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 4\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 5\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 6\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 7\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 8\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 9\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 10\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0128 - mse: 0.0128 - val_loss: 0.0131 - val_mse: 0.0131\n"
     ]
    }
   ],
   "source": [
    "run,r2scores,msescores,maescores,pearsoncorr = [],[],[],[],[]\n",
    "for i in range(10):\n",
    "    print('\\nRun = {}'.format(i+1))\n",
    "    X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "    model = MLP2(np.shape(X_train)[1],np.shape(y_train)[1],16)\n",
    "    model.fit(np.array(X_train), np.array(y_train), epochs=10, batch_size=16, validation_split=0.2, shuffle=True)\n",
    "    y_pred = model.predict(np.array(X_test))\n",
    "    \n",
    "    #results\n",
    "    run.append(i+1)\n",
    "    \n",
    "    actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    r2scores.append(r2_score(actual_mean, pred_mean))\n",
    "    msescores.append(mean_squared_error(np.array(y_test), y_pred))\n",
    "    maescores.append(mean_absolute_error(np.array(y_test), y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 42,
     "status": "ok",
     "timestamp": 1647772685188,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "EdEHeH8_5u-o",
    "outputId": "bcd70bf6-eda1-4aee-c7e1-94544001365c",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-70ecfbea-1392-43e8-98b0-81f120805f3d\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>#Run</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.684423</td>\n",
       "      <td>0.013254</td>\n",
       "      <td>0.045950</td>\n",
       "      <td>0.695382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.660864</td>\n",
       "      <td>0.013298</td>\n",
       "      <td>0.046047</td>\n",
       "      <td>0.687590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.625262</td>\n",
       "      <td>0.013381</td>\n",
       "      <td>0.047205</td>\n",
       "      <td>0.683113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "      <td>0.684754</td>\n",
       "      <td>0.013256</td>\n",
       "      <td>0.046328</td>\n",
       "      <td>0.692560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>0.692972</td>\n",
       "      <td>0.013249</td>\n",
       "      <td>0.046704</td>\n",
       "      <td>0.700950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.683927</td>\n",
       "      <td>0.013250</td>\n",
       "      <td>0.046220</td>\n",
       "      <td>0.695313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7.0</td>\n",
       "      <td>0.692177</td>\n",
       "      <td>0.013239</td>\n",
       "      <td>0.047312</td>\n",
       "      <td>0.705137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8.0</td>\n",
       "      <td>0.684794</td>\n",
       "      <td>0.013268</td>\n",
       "      <td>0.046858</td>\n",
       "      <td>0.700424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9.0</td>\n",
       "      <td>0.670181</td>\n",
       "      <td>0.013286</td>\n",
       "      <td>0.046796</td>\n",
       "      <td>0.692998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10.0</td>\n",
       "      <td>0.669955</td>\n",
       "      <td>0.013292</td>\n",
       "      <td>0.045688</td>\n",
       "      <td>0.697035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>5.5</td>\n",
       "      <td>0.674931</td>\n",
       "      <td>0.013277</td>\n",
       "      <td>0.046511</td>\n",
       "      <td>0.695050</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-70ecfbea-1392-43e8-98b0-81f120805f3d')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-70ecfbea-1392-43e8-98b0-81f120805f3d button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-70ecfbea-1392-43e8-98b0-81f120805f3d');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "     #Run   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     1.0  0.684423  0.013254  0.045950            0.695382\n",
       "1     2.0  0.660864  0.013298  0.046047            0.687590\n",
       "2     3.0  0.625262  0.013381  0.047205            0.683113\n",
       "3     4.0  0.684754  0.013256  0.046328            0.692560\n",
       "4     5.0  0.692972  0.013249  0.046704            0.700950\n",
       "5     6.0  0.683927  0.013250  0.046220            0.695313\n",
       "6     7.0  0.692177  0.013239  0.047312            0.705137\n",
       "7     8.0  0.684794  0.013268  0.046858            0.700424\n",
       "8     9.0  0.670181  0.013286  0.046796            0.692998\n",
       "9    10.0  0.669955  0.013292  0.045688            0.697035\n",
       "avg   5.5  0.674931  0.013277  0.046511            0.695050"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(run,r2scores,msescores,maescores,pearsoncorr))\n",
    "result.columns = ['#Run','R2Score','MSE','MAE', 'Correlation Coeff.']\n",
    "result.loc[len(run)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(run):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "executionInfo": {
     "elapsed": 37,
     "status": "ok",
     "timestamp": 1647772685189,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "xU2mV9qO5u-o"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/DNN_10PercentDropout_10runs.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "executionInfo": {
     "elapsed": 36,
     "status": "ok",
     "timestamp": 1647772685190,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "SQM1Qm4dpwAg",
    "outputId": "e524796d-8d5f-408b-a83f-efcb889c03ab"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_39\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dropout_27 (Dropout)        (None, 143)               0         \n",
      "                                                                 \n",
      " dense_117 (Dense)           (None, 130)               18720     \n",
      "                                                                 \n",
      " dropout_28 (Dropout)        (None, 130)               0         \n",
      "                                                                 \n",
      " dense_118 (Dense)           (None, 75)                9825      \n",
      "                                                                 \n",
      " dropout_29 (Dropout)        (None, 75)                0         \n",
      "                                                                 \n",
      " dense_119 (Dense)           (None, 66)                5016      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 33,561\n",
      "Trainable params: 33,561\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.summary()\n",
    "plot_model(model, show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "aYWTvMiy5u-s"
   },
   "source": [
    "**DNN with 10% dropout at input layer and 20% dropout at hidden layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "executionInfo": {
     "elapsed": 15,
     "status": "ok",
     "timestamp": 1647772685190,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "VKYOUWYH5u-s"
   },
   "outputs": [],
   "source": [
    "def MLP3(x_shape,y_shape,rnd):\n",
    "    model = Sequential()\n",
    "    model.add(Dropout(0.1,seed=rnd))\n",
    "    model.add(Dense(130, input_dim=x_shape, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dropout(0.2,seed=rnd))\n",
    "    model.add(Dense(75, kernel_initializer='uniform', activation=\"relu\"))\n",
    "    model.add(Dropout(0.2,seed=rnd))\n",
    "    model.add(Dense(y_shape, kernel_initializer='uniform', activation=\"linear\"))\n",
    "    model.compile(optimizer='adam',loss='mean_squared_error', metrics=['mse'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 155548,
     "status": "ok",
     "timestamp": 1647772840724,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "TpbvG-Da5u-t",
    "outputId": "fd3904f6-9623-4fbf-d778-bf24b8542fbb",
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Run = 1\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 2\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0134 - val_mse: 0.0134\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 3\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0135 - mse: 0.0135 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "\n",
      "Run = 4\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0133 - val_mse: 0.0133\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "\n",
      "Run = 5\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 6\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 7\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0129 - val_mse: 0.0129\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 8\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "\n",
      "Run = 9\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0132 - mse: 0.0132 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "\n",
      "Run = 10\n",
      "Epoch 1/10\n",
      "416/416 [==============================] - 2s 3ms/step - loss: 0.0136 - mse: 0.0136 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 2/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0134 - mse: 0.0134 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 3/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0133 - mse: 0.0133 - val_loss: 0.0132 - val_mse: 0.0132\n",
      "Epoch 4/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0131 - val_mse: 0.0131\n",
      "Epoch 5/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0131 - mse: 0.0131 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 6/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 7/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0130 - mse: 0.0130 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 8/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 9/10\n",
      "416/416 [==============================] - 1s 3ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n",
      "Epoch 10/10\n",
      "416/416 [==============================] - 1s 2ms/step - loss: 0.0129 - mse: 0.0129 - val_loss: 0.0130 - val_mse: 0.0130\n"
     ]
    }
   ],
   "source": [
    "run,r2scores,msescores,maescores,pearsoncorr = [],[],[],[],[]\n",
    "for i in range(10):\n",
    "    print('\\nRun = {}'.format(i+1))\n",
    "    X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "    model = MLP3(np.shape(X_train)[1],np.shape(y_train)[1],16)\n",
    "    model.fit(np.array(X_train), np.array(y_train), epochs=10, batch_size=16, validation_split=0.2, shuffle=True)\n",
    "    y_pred = model.predict(np.array(X_test))\n",
    "    \n",
    "    #results\n",
    "    run.append(i)\n",
    "    \n",
    "    actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    r2scores.append(r2_score(actual_mean, pred_mean))\n",
    "    msescores.append(mean_squared_error(np.array(y_test), y_pred))\n",
    "    maescores.append(mean_absolute_error(np.array(y_test), y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 1098,
     "status": "ok",
     "timestamp": 1647772841811,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "m1bE-rWl5u-t",
    "outputId": "13d14b78-cccc-411f-9c3c-e9e003917dd3"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-cf79b6fc-59fd-4541-afb3-4c511279dbe4\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>#Run</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.694834</td>\n",
       "      <td>0.013232</td>\n",
       "      <td>0.045610</td>\n",
       "      <td>0.695200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.675348</td>\n",
       "      <td>0.013269</td>\n",
       "      <td>0.046012</td>\n",
       "      <td>0.680798</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.680776</td>\n",
       "      <td>0.013265</td>\n",
       "      <td>0.045308</td>\n",
       "      <td>0.695753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.693657</td>\n",
       "      <td>0.013227</td>\n",
       "      <td>0.045726</td>\n",
       "      <td>0.700665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.0</td>\n",
       "      <td>0.667418</td>\n",
       "      <td>0.013287</td>\n",
       "      <td>0.046609</td>\n",
       "      <td>0.699576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5.0</td>\n",
       "      <td>0.693952</td>\n",
       "      <td>0.013235</td>\n",
       "      <td>0.045754</td>\n",
       "      <td>0.695241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.671211</td>\n",
       "      <td>0.013288</td>\n",
       "      <td>0.046087</td>\n",
       "      <td>0.698546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7.0</td>\n",
       "      <td>0.640107</td>\n",
       "      <td>0.013345</td>\n",
       "      <td>0.046549</td>\n",
       "      <td>0.691400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8.0</td>\n",
       "      <td>0.683758</td>\n",
       "      <td>0.013248</td>\n",
       "      <td>0.044720</td>\n",
       "      <td>0.691739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9.0</td>\n",
       "      <td>0.673826</td>\n",
       "      <td>0.013272</td>\n",
       "      <td>0.045921</td>\n",
       "      <td>0.682522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>4.5</td>\n",
       "      <td>0.677489</td>\n",
       "      <td>0.013267</td>\n",
       "      <td>0.045830</td>\n",
       "      <td>0.693144</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cf79b6fc-59fd-4541-afb3-4c511279dbe4')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-cf79b6fc-59fd-4541-afb3-4c511279dbe4 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-cf79b6fc-59fd-4541-afb3-4c511279dbe4');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "     #Run   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     0.0  0.694834  0.013232  0.045610            0.695200\n",
       "1     1.0  0.675348  0.013269  0.046012            0.680798\n",
       "2     2.0  0.680776  0.013265  0.045308            0.695753\n",
       "3     3.0  0.693657  0.013227  0.045726            0.700665\n",
       "4     4.0  0.667418  0.013287  0.046609            0.699576\n",
       "5     5.0  0.693952  0.013235  0.045754            0.695241\n",
       "6     6.0  0.671211  0.013288  0.046087            0.698546\n",
       "7     7.0  0.640107  0.013345  0.046549            0.691400\n",
       "8     8.0  0.683758  0.013248  0.044720            0.691739\n",
       "9     9.0  0.673826  0.013272  0.045921            0.682522\n",
       "avg   4.5  0.677489  0.013267  0.045830            0.693144"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(run,r2scores,msescores,maescores,pearsoncorr))\n",
    "result.columns = ['#Run','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(run)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(run):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "executionInfo": {
     "elapsed": 37,
     "status": "ok",
     "timestamp": 1647772841811,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "-IQdeDBN5u-u"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/DNN_10Percent20PercentDropout_10runs.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "executionInfo": {
     "elapsed": 34,
     "status": "ok",
     "timestamp": 1647772841812,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "l4OyKXV-4yjk",
    "outputId": "d783a714-441d-43e6-d247-8df492fc9545"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_49\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dropout_57 (Dropout)        (None, 143)               0         \n",
      "                                                                 \n",
      " dense_147 (Dense)           (None, 130)               18720     \n",
      "                                                                 \n",
      " dropout_58 (Dropout)        (None, 130)               0         \n",
      "                                                                 \n",
      " dense_148 (Dense)           (None, 75)                9825      \n",
      "                                                                 \n",
      " dropout_59 (Dropout)        (None, 75)                0         \n",
      "                                                                 \n",
      " dense_149 (Dense)           (None, 66)                5016      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 33,561\n",
      "Trainable params: 33,561\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.summary()\n",
    "plot_model(model, show_shapes=True, show_layer_names=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QyxLod3yDJF3"
   },
   "source": [
    "# Benchmarking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "executionInfo": {
     "elapsed": 725,
     "status": "ok",
     "timestamp": 1647772948758,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "5VBcRZR1B0Wg"
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RXMuqJrEDWWG"
   },
   "source": [
    "# Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "executionInfo": {
     "elapsed": 7878,
     "status": "ok",
     "timestamp": 1647772968195,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "TOSzLIKjDSYu"
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import r2_score \n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "run,r2scores_lr,msescores_lr,maescores_lr,pearsoncorr = [],[],[],[],[]\n",
    "\n",
    "for i in range(10):\n",
    "    X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "    linear_Regr = LinearRegression(normalize=True)\n",
    "    linear_Regr.fit(X_train, y_train)\n",
    "    y_pred = linear_Regr.predict(X_test)\n",
    "    \n",
    "    run.append(i+1)\n",
    "    actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    r2scores_lr.append(r2_score(actual_mean, pred_mean))\n",
    "    msescores_lr.append(mean_squared_error(y_test, y_pred))\n",
    "    maescores_lr.append(mean_absolute_error(y_test, y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 40,
     "status": "ok",
     "timestamp": 1647772968197,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "VHMVPkgd5u-y",
    "outputId": "b0c82a0d-0d09-4b6e-d3c4-2578b0c1e9d3"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-099dd016-5043-487d-acc4-9f9cce787de8\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>#Run</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10.0</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>5.5</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638879</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-099dd016-5043-487d-acc4-9f9cce787de8')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-099dd016-5043-487d-acc4-9f9cce787de8 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-099dd016-5043-487d-acc4-9f9cce787de8');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "     #Run   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     1.0  0.556212  0.013722  0.047631            0.638879\n",
       "1     2.0  0.556212  0.013722  0.047631            0.638879\n",
       "2     3.0  0.556212  0.013722  0.047631            0.638879\n",
       "3     4.0  0.556212  0.013722  0.047631            0.638879\n",
       "4     5.0  0.556212  0.013722  0.047631            0.638879\n",
       "5     6.0  0.556212  0.013722  0.047631            0.638879\n",
       "6     7.0  0.556212  0.013722  0.047631            0.638879\n",
       "7     8.0  0.556212  0.013722  0.047631            0.638879\n",
       "8     9.0  0.556212  0.013722  0.047631            0.638879\n",
       "9    10.0  0.556212  0.013722  0.047631            0.638879\n",
       "avg   5.5  0.556212  0.013722  0.047631            0.638879"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(run,r2scores_lr,msescores_lr,maescores_lr,pearsoncorr))\n",
    "result.columns = ['#Run','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(run)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(run):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "executionInfo": {
     "elapsed": 36,
     "status": "ok",
     "timestamp": 1647772968198,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "XDmh9maH5u-z"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/LinearRegression_10runs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "l9d3EKNHF4Ap"
   },
   "source": [
    "# Lasso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 11898,
     "status": "ok",
     "timestamp": 1647772989829,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "9jzWOKAZFwqi",
    "outputId": "368e3fcb-79da-441f-80e8-52ca624b5ff3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 1e-05\n",
      "Working with alpha = 5e-05\n",
      "Working with alpha = 0.0001\n",
      "Working with alpha = 0.0005\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.001\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.005\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.01\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.05\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha = 0.5\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/scipy/stats/stats.py:3508: PearsonRConstantInputWarning: An input array is constant; the correlation coefficent is not defined.\n",
      "  warnings.warn(PearsonRConstantInputWarning())\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import Lasso\n",
    "\n",
    "alpha,ls_r2,ls_mse,ls_mae,pearsoncorr = [],[],[],[],[]\n",
    "X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "\n",
    "#for alp in [0.01,0.05,0.1,0.2,0.5,1,2,2.5,5,10]:\n",
    "for alp in [0.00001,0.00005,0.0001,0.0005,0.001,0.005,0.01,0.05,0.1,0.5]:\n",
    "    print('Working with alpha =',alp)\n",
    "    Lasso_Regr = Lasso(alpha=alp, normalize=True)\n",
    "    Lasso_Regr.fit(X_train,y_train)\n",
    "    y_pred = Lasso_Regr.predict(X_test)\n",
    "    \n",
    "    alpha.append(alp)\n",
    "    \n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    ls_r2.append(r2_score(actual_mean, pred_mean))\n",
    "    ls_mse.append(mean_squared_error(y_test, y_pred))\n",
    "    ls_mae.append(mean_absolute_error(y_test, y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 24,
     "status": "ok",
     "timestamp": 1647772989831,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "QbLFjfai5u-2",
    "outputId": "efd1d3fc-5ea7-4546-dd48-6f9306976876"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-1bee8611-ab42-4f5d-ba87-aec2e57ac4c2\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Alpha</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.547076</td>\n",
       "      <td>0.013631</td>\n",
       "      <td>0.047034</td>\n",
       "      <td>0.633291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.410086</td>\n",
       "      <td>0.013860</td>\n",
       "      <td>0.048876</td>\n",
       "      <td>0.637886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000100</td>\n",
       "      <td>0.172365</td>\n",
       "      <td>0.014328</td>\n",
       "      <td>0.050911</td>\n",
       "      <td>0.584811</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000500</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.001000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.005000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.010000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.050000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.100000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>-0.001081</td>\n",
       "      <td>0.014704</td>\n",
       "      <td>0.052136</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>0.066666</td>\n",
       "      <td>0.112196</td>\n",
       "      <td>0.014475</td>\n",
       "      <td>0.051177</td>\n",
       "      <td>0.618662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1bee8611-ab42-4f5d-ba87-aec2e57ac4c2')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-1bee8611-ab42-4f5d-ba87-aec2e57ac4c2 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-1bee8611-ab42-4f5d-ba87-aec2e57ac4c2');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "        Alpha   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0    0.000010  0.547076  0.013631  0.047034            0.633291\n",
       "1    0.000050  0.410086  0.013860  0.048876            0.637886\n",
       "2    0.000100  0.172365  0.014328  0.050911            0.584811\n",
       "3    0.000500 -0.001081  0.014704  0.052136                 NaN\n",
       "4    0.001000 -0.001081  0.014704  0.052136                 NaN\n",
       "5    0.005000 -0.001081  0.014704  0.052136                 NaN\n",
       "6    0.010000 -0.001081  0.014704  0.052136                 NaN\n",
       "7    0.050000 -0.001081  0.014704  0.052136                 NaN\n",
       "8    0.100000 -0.001081  0.014704  0.052136                 NaN\n",
       "9    0.500000 -0.001081  0.014704  0.052136                 NaN\n",
       "avg  0.066666  0.112196  0.014475  0.051177            0.618662"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(alpha,ls_r2,ls_mse,ls_mae,pearsoncorr))\n",
    "result.columns = ['Alpha','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(alpha)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(alpha):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "executionInfo": {
     "elapsed": 19,
     "status": "ok",
     "timestamp": 1647772989832,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "UEhTIj2a5u-2"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/LassoRegression_10runs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gDukM2LWjsGO"
   },
   "source": [
    "# Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3555,
     "status": "ok",
     "timestamp": 1647772993369,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "NXBT6ZilHWZL",
    "outputId": "8ca489a9-6035-439e-d833-55c9fc44c474"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Working with alpha =  1e-05\n",
      "Working with alpha =  5e-05\n",
      "Working with alpha =  0.0001\n",
      "Working with alpha =  0.0005\n",
      "Working with alpha =  0.001\n",
      "Working with alpha =  0.005\n",
      "Working with alpha =  0.01\n",
      "Working with alpha =  0.05\n",
      "Working with alpha =  0.1\n",
      "Working with alpha =  0.5\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import Ridge\n",
    "\n",
    "alpha,rg_r2,rg_mse,rg_mae,pearsoncorr = [],[],[],[],[]\n",
    "X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "\n",
    "for alp in [0.00001,0.00005,0.0001,0.0005,0.001,0.005,0.01,0.05,0.1,0.5]:\n",
    "    print('Working with alpha = ',alp)\n",
    "    Ridge_Regr = Ridge(alpha=alp, normalize=True)\n",
    "    Ridge_Regr.fit(X_train, y_train)\n",
    "    y_pred = Ridge_Regr.predict(X_test)\n",
    "    \n",
    "    alpha.append(alp)\n",
    "    \n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    rg_r2.append(r2_score(actual_mean, pred_mean))\n",
    "    rg_mse.append(mean_squared_error(y_test, y_pred))\n",
    "    rg_mae.append(mean_absolute_error(y_test, y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1647772993370,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "1XFxEgCr5u-7",
    "outputId": "52085070-5b87-4d82-9a36-9d8c24ef3035"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-ec310da1-4c3e-47fa-bd0e-f803b505fc0a\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Alpha</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000010</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000050</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000100</td>\n",
       "      <td>0.556212</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047631</td>\n",
       "      <td>0.638877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000500</td>\n",
       "      <td>0.556213</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>0.047630</td>\n",
       "      <td>0.638872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.001000</td>\n",
       "      <td>0.556214</td>\n",
       "      <td>0.013721</td>\n",
       "      <td>0.047628</td>\n",
       "      <td>0.638866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.005000</td>\n",
       "      <td>0.556221</td>\n",
       "      <td>0.013719</td>\n",
       "      <td>0.047614</td>\n",
       "      <td>0.638814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.010000</td>\n",
       "      <td>0.556228</td>\n",
       "      <td>0.013717</td>\n",
       "      <td>0.047596</td>\n",
       "      <td>0.638750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.050000</td>\n",
       "      <td>0.556188</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>0.047473</td>\n",
       "      <td>0.638233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.555934</td>\n",
       "      <td>0.013680</td>\n",
       "      <td>0.047350</td>\n",
       "      <td>0.637589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.549376</td>\n",
       "      <td>0.013609</td>\n",
       "      <td>0.047015</td>\n",
       "      <td>0.632919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>0.066666</td>\n",
       "      <td>0.555501</td>\n",
       "      <td>0.013703</td>\n",
       "      <td>0.047520</td>\n",
       "      <td>0.638068</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ec310da1-4c3e-47fa-bd0e-f803b505fc0a')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-ec310da1-4c3e-47fa-bd0e-f803b505fc0a button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-ec310da1-4c3e-47fa-bd0e-f803b505fc0a');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "        Alpha   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0    0.000010  0.556212  0.013722  0.047631            0.638878\n",
       "1    0.000050  0.556212  0.013722  0.047631            0.638878\n",
       "2    0.000100  0.556212  0.013722  0.047631            0.638877\n",
       "3    0.000500  0.556213  0.013722  0.047630            0.638872\n",
       "4    0.001000  0.556214  0.013721  0.047628            0.638866\n",
       "5    0.005000  0.556221  0.013719  0.047614            0.638814\n",
       "6    0.010000  0.556228  0.013717  0.047596            0.638750\n",
       "7    0.050000  0.556188  0.013699  0.047473            0.638233\n",
       "8    0.100000  0.555934  0.013680  0.047350            0.637589\n",
       "9    0.500000  0.549376  0.013609  0.047015            0.632919\n",
       "avg  0.066666  0.555501  0.013703  0.047520            0.638068"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(alpha,rg_r2,rg_mse,rg_mae,pearsoncorr))\n",
    "result.columns = ['Alpha','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(alpha)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(alpha):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "executionInfo": {
     "elapsed": 11,
     "status": "ok",
     "timestamp": 1647772993372,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "EDNEjlWt5u-7"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/RidgeRegression_10runs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Kav5r__yqC1V"
   },
   "source": [
    "# k-NN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 13683,
     "status": "ok",
     "timestamp": 1647773007045,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "8GxLmGfkpy3w",
    "outputId": "54eb821d-0665-4af9-ebd2-76bab56962de"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "k= 1\n",
      "k= 5\n",
      "k= 9\n",
      "k= 11\n",
      "k= 15\n",
      "k= 19\n",
      "k= 21\n",
      "k= 25\n",
      "k= 29\n",
      "k= 31\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "\n",
    "knn,knn_r2,knn_mse,knn_mae=[],[],[],[]\n",
    "X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "\n",
    "for k in [1,5,9,11,15,19,21,25,29,31]:\n",
    "    print('k=',k)\n",
    "    knn_Regr = KNeighborsRegressor(n_neighbors=k, n_jobs=-1)\n",
    "    knn_Regr.fit(X_train, y_train)\n",
    "    y_pred = knn_Regr.predict(X_test)\n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    knn.append(k)\n",
    "    knn_r2.append(r2_score(actual_mean, pred_mean))\n",
    "    knn_mse.append(mean_squared_error(y_test, y_pred))\n",
    "    knn_mae.append(mean_absolute_error(y_test, y_pred))\n",
    "\n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 23,
     "status": "ok",
     "timestamp": 1647773007046,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "EX60668T5u_D",
    "outputId": "10b34703-d6d8-4003-ab27-4ad943ba3d46"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-a8609eec-1fa3-4ddd-8595-0cf1d475d966\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>k</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.384715</td>\n",
       "      <td>0.021957</td>\n",
       "      <td>0.041695</td>\n",
       "      <td>0.638878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.0</td>\n",
       "      <td>0.578562</td>\n",
       "      <td>0.014969</td>\n",
       "      <td>0.039650</td>\n",
       "      <td>0.638878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9.0</td>\n",
       "      <td>0.594137</td>\n",
       "      <td>0.014267</td>\n",
       "      <td>0.039948</td>\n",
       "      <td>0.638877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11.0</td>\n",
       "      <td>0.605558</td>\n",
       "      <td>0.014117</td>\n",
       "      <td>0.040292</td>\n",
       "      <td>0.638872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15.0</td>\n",
       "      <td>0.609513</td>\n",
       "      <td>0.013942</td>\n",
       "      <td>0.040930</td>\n",
       "      <td>0.638866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>19.0</td>\n",
       "      <td>0.610651</td>\n",
       "      <td>0.013855</td>\n",
       "      <td>0.041709</td>\n",
       "      <td>0.638814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>21.0</td>\n",
       "      <td>0.616576</td>\n",
       "      <td>0.013811</td>\n",
       "      <td>0.041992</td>\n",
       "      <td>0.638750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>25.0</td>\n",
       "      <td>0.620453</td>\n",
       "      <td>0.013744</td>\n",
       "      <td>0.042578</td>\n",
       "      <td>0.638233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>29.0</td>\n",
       "      <td>0.618618</td>\n",
       "      <td>0.013715</td>\n",
       "      <td>0.043194</td>\n",
       "      <td>0.637589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>31.0</td>\n",
       "      <td>0.617953</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>0.043449</td>\n",
       "      <td>0.632919</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>16.6</td>\n",
       "      <td>0.585674</td>\n",
       "      <td>0.014808</td>\n",
       "      <td>0.041544</td>\n",
       "      <td>0.638068</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a8609eec-1fa3-4ddd-8595-0cf1d475d966')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-a8609eec-1fa3-4ddd-8595-0cf1d475d966 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-a8609eec-1fa3-4ddd-8595-0cf1d475d966');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "        k   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     1.0  0.384715  0.021957  0.041695            0.638878\n",
       "1     5.0  0.578562  0.014969  0.039650            0.638878\n",
       "2     9.0  0.594137  0.014267  0.039948            0.638877\n",
       "3    11.0  0.605558  0.014117  0.040292            0.638872\n",
       "4    15.0  0.609513  0.013942  0.040930            0.638866\n",
       "5    19.0  0.610651  0.013855  0.041709            0.638814\n",
       "6    21.0  0.616576  0.013811  0.041992            0.638750\n",
       "7    25.0  0.620453  0.013744  0.042578            0.638233\n",
       "8    29.0  0.618618  0.013715  0.043194            0.637589\n",
       "9    31.0  0.617953  0.013699  0.043449            0.632919\n",
       "avg  16.6  0.585674  0.014808  0.041544            0.638068"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(knn,knn_r2,knn_mse,knn_mae,pearsoncorr))\n",
    "result.columns = ['k','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(knn)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(knn):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "executionInfo": {
     "elapsed": 8,
     "status": "ok",
     "timestamp": 1647773007047,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "rK64h62V5u_E"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/kNNRegression_10runs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CALVTtl3m1Bh"
   },
   "source": [
    "# Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 830352,
     "status": "ok",
     "timestamp": 1647773837392,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "lM5zTktSmfz4",
    "outputId": "755006b5-435a-4792-b566-69b272fc5a9b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "estimators =  10\n",
      "estimators =  20\n",
      "estimators =  30\n",
      "estimators =  40\n",
      "estimators =  50\n",
      "estimators =  60\n",
      "estimators =  70\n",
      "estimators =  80\n",
      "estimators =  90\n",
      "estimators =  100\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "estimators,rf_r2,rf_mse,rf_mae,pearsoncorr = [],[],[],[],[]\n",
    "X_train,X_test,y_train,y_test = train_test_split(X_data,y_data,test_size = 0.2,shuffle=True,random_state=16)\n",
    "actual_mean = pd.DataFrame(y_test.mean(axis=1))\n",
    "\n",
    "for est in [10,20,30,40,50,60,70,80,90,100]:\n",
    "    print('estimators = ',est)\n",
    "    rf_Regr = RandomForestRegressor(n_estimators=est, n_jobs=-1)\n",
    "    rf_Regr.fit(X_train, y_train)\n",
    "    y_pred = rf_Regr.predict(X_test)\n",
    "    \n",
    "    estimators.append(est)\n",
    "    \n",
    "    pred_mean = pd.DataFrame(y_pred.mean(axis=1))\n",
    "    rf_r2.append(r2_score(actual_mean, pred_mean))\n",
    "    rf_mse.append(mean_squared_error(y_test, y_pred))\n",
    "    rf_mae.append(mean_absolute_error(y_test, y_pred))\n",
    "    \n",
    "    dist_orig = np.square(euclidean_distances(y_test, y_test)).flatten()\n",
    "    dist_pred = np.square(euclidean_distances(y_pred, y_pred)).flatten()\n",
    "    corr, _ = pearsonr(dist_orig, dist_pred)\n",
    "    pearsoncorr.append(corr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 394
    },
    "executionInfo": {
     "elapsed": 29,
     "status": "ok",
     "timestamp": 1647773837398,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "_i6My7wY5u-_",
    "outputId": "d8619601-9488-4330-b397-00d0296cb653"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-276aba5d-6905-4f79-8bb5-19151080e0f0\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Alpha</th>\n",
       "      <th>R2Score</th>\n",
       "      <th>MSE</th>\n",
       "      <th>MAE</th>\n",
       "      <th>Correlation Coeff.</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10.0</td>\n",
       "      <td>0.618462</td>\n",
       "      <td>0.014650</td>\n",
       "      <td>0.045127</td>\n",
       "      <td>0.667525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20.0</td>\n",
       "      <td>0.658857</td>\n",
       "      <td>0.013930</td>\n",
       "      <td>0.044783</td>\n",
       "      <td>0.686457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30.0</td>\n",
       "      <td>0.655810</td>\n",
       "      <td>0.013764</td>\n",
       "      <td>0.044924</td>\n",
       "      <td>0.684941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>40.0</td>\n",
       "      <td>0.651116</td>\n",
       "      <td>0.013648</td>\n",
       "      <td>0.044630</td>\n",
       "      <td>0.680590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>50.0</td>\n",
       "      <td>0.645630</td>\n",
       "      <td>0.013604</td>\n",
       "      <td>0.044537</td>\n",
       "      <td>0.681242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60.0</td>\n",
       "      <td>0.660290</td>\n",
       "      <td>0.013536</td>\n",
       "      <td>0.044428</td>\n",
       "      <td>0.684913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>70.0</td>\n",
       "      <td>0.653899</td>\n",
       "      <td>0.013522</td>\n",
       "      <td>0.044544</td>\n",
       "      <td>0.676631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>80.0</td>\n",
       "      <td>0.650742</td>\n",
       "      <td>0.013502</td>\n",
       "      <td>0.044525</td>\n",
       "      <td>0.674176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>90.0</td>\n",
       "      <td>0.659886</td>\n",
       "      <td>0.013462</td>\n",
       "      <td>0.044434</td>\n",
       "      <td>0.685219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>100.0</td>\n",
       "      <td>0.660560</td>\n",
       "      <td>0.013447</td>\n",
       "      <td>0.044431</td>\n",
       "      <td>0.684871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>avg</th>\n",
       "      <td>55.0</td>\n",
       "      <td>0.651525</td>\n",
       "      <td>0.013706</td>\n",
       "      <td>0.044636</td>\n",
       "      <td>0.680656</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-276aba5d-6905-4f79-8bb5-19151080e0f0')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-276aba5d-6905-4f79-8bb5-19151080e0f0 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-276aba5d-6905-4f79-8bb5-19151080e0f0');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "     Alpha   R2Score       MSE       MAE  Correlation Coeff.\n",
       "0     10.0  0.618462  0.014650  0.045127            0.667525\n",
       "1     20.0  0.658857  0.013930  0.044783            0.686457\n",
       "2     30.0  0.655810  0.013764  0.044924            0.684941\n",
       "3     40.0  0.651116  0.013648  0.044630            0.680590\n",
       "4     50.0  0.645630  0.013604  0.044537            0.681242\n",
       "5     60.0  0.660290  0.013536  0.044428            0.684913\n",
       "6     70.0  0.653899  0.013522  0.044544            0.676631\n",
       "7     80.0  0.650742  0.013502  0.044525            0.674176\n",
       "8     90.0  0.659886  0.013462  0.044434            0.685219\n",
       "9    100.0  0.660560  0.013447  0.044431            0.684871\n",
       "avg   55.0  0.651525  0.013706  0.044636            0.680656"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.DataFrame(zip(estimators,rf_r2,rf_mse,rf_mae,pearsoncorr))\n",
    "result.columns = ['Alpha','R2Score','MSE','MAE','Correlation Coeff.']\n",
    "result.loc[len(estimators)] = result.mean(numeric_only=True, axis=0)\n",
    "result.rename(index={len(estimators):'avg'},inplace=True)\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "executionInfo": {
     "elapsed": 23,
     "status": "ok",
     "timestamp": 1647773837399,
     "user": {
      "displayName": "Dibyendu B. Seal AKCSIT, CU",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GhbHn18MudnPFuAJzhuld5U2wPbe7E9gBEbrOAO=s64",
      "userId": "00429258219838613002"
     },
     "user_tz": -330
    },
    "id": "ku-f0Bdn5u_A"
   },
   "outputs": [],
   "source": [
    "result.to_csv(dir+'Results/EMTAB7606/RFRegression_10runs.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "vOnR2EHp7IYZ"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "EMTAB7606_scRNAseqTimeSeries.ipynb",
   "provenance": [
    {
     "file_id": "1mvJ_jyvrWXkBesVasIcn2dKcIcyYyahT",
     "timestamp": 1647543809972
    }
   ],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}