_nn_regs_stats.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e20ff980",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn\n",
    "import joblib\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "%store -r X_features_full\n",
    "%store -r X_features_poly_full\n",
    "%store -r y_labels_full\n",
    "\n",
    "%store -r totaldf_full\n",
    "X_features = X_features_full\n",
    "X_features_poly = X_features_poly_full\n",
    "y_labels = y_labels_full\n",
    "totaldf = totaldf_full"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8fe55a25",
   "metadata": {},
   "outputs": [],
   "source": [
    "seq_model = keras.models.load_model('D:\\python\\\\full data\\\\first')\n",
    "func_model = keras.models.load_model('D:\\python\\\\full data\\\\second')\n",
    "rnd_model = keras.models.load_model('D:\\python\\\\full data\\\\third')\n",
    "poly_model = keras.models.load_model('D:\\python\\\\full data\\\\fourth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "79f69a51",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "72/72 [==============================] - 0s 1ms/step\n",
      "72/72 [==============================] - 0s 2ms/step\n",
      "72/72 [==============================] - 0s 2ms/step\n",
      "72/72 [==============================] - 0s 2ms/step\n"
     ]
    }
   ],
   "source": [
    "totaldf.insert(7,'sequential', (np.exp(np.power(seq_model.predict(X_features[2]),2)) -0.0012))\n",
    "totaldf.insert(8,'functional',(np.exp(np.power(func_model.predict(X_features[2]),2)) -0.0012))\n",
    "totaldf.insert(9,'3layerdeep',(np.exp(np.power(rnd_model.predict(X_features[2]),2)) -0.0012))\n",
    "totaldf.insert(10,'poly',(np.exp(np.power(poly_model.predict(X_features[2]),2)) -0.0012))\n",
    "totaldf.sequential = totaldf.sequential*totaldf.μob\n",
    "totaldf.functional = totaldf.functional*totaldf.μob\n",
    "totaldf['3layerdeep'] = totaldf['3layerdeep']*totaldf.μob\n",
    "totaldf.poly = totaldf.poly*totaldf.μob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7837cccd",
   "metadata": {},
   "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>Pb</th>\n",
       "      <th>P</th>\n",
       "      <th>ΔP</th>\n",
       "      <th>μob</th>\n",
       "      <th>μ</th>\n",
       "      <th>Rsob</th>\n",
       "      <th>API</th>\n",
       "      <th>sequential</th>\n",
       "      <th>functional</th>\n",
       "      <th>3layerdeep</th>\n",
       "      <th>poly</th>\n",
       "      <th>Tr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2200.008903</td>\n",
       "      <td>2200.008903</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>478.607857</td>\n",
       "      <td>32.545608</td>\n",
       "      <td>0.998470</td>\n",
       "      <td>0.998585</td>\n",
       "      <td>0.998694</td>\n",
       "      <td>0.998792</td>\n",
       "      <td>336.150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2200.008903</td>\n",
       "      <td>2348.166747</td>\n",
       "      <td>148.157844</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>1.007916</td>\n",
       "      <td>478.607857</td>\n",
       "      <td>32.545608</td>\n",
       "      <td>1.007414</td>\n",
       "      <td>1.007508</td>\n",
       "      <td>1.008885</td>\n",
       "      <td>1.008426</td>\n",
       "      <td>336.150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2200.008903</td>\n",
       "      <td>2496.324590</td>\n",
       "      <td>296.315687</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>1.017489</td>\n",
       "      <td>478.607857</td>\n",
       "      <td>32.545608</td>\n",
       "      <td>1.015576</td>\n",
       "      <td>1.018905</td>\n",
       "      <td>1.015306</td>\n",
       "      <td>1.018305</td>\n",
       "      <td>336.150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2200.008903</td>\n",
       "      <td>2644.482434</td>\n",
       "      <td>444.473531</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>1.027342</td>\n",
       "      <td>478.607857</td>\n",
       "      <td>32.545608</td>\n",
       "      <td>1.027691</td>\n",
       "      <td>1.027661</td>\n",
       "      <td>1.020632</td>\n",
       "      <td>1.033516</td>\n",
       "      <td>336.150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2200.008903</td>\n",
       "      <td>2792.640278</td>\n",
       "      <td>592.631375</td>\n",
       "      <td>0.998677</td>\n",
       "      <td>1.037429</td>\n",
       "      <td>478.607857</td>\n",
       "      <td>32.545608</td>\n",
       "      <td>1.039029</td>\n",
       "      <td>1.038241</td>\n",
       "      <td>1.027324</td>\n",
       "      <td>1.039544</td>\n",
       "      <td>336.150</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>3910.006923</td>\n",
       "      <td>5560.005012</td>\n",
       "      <td>1649.998089</td>\n",
       "      <td>0.125500</td>\n",
       "      <td>0.155127</td>\n",
       "      <td>3229.949353</td>\n",
       "      <td>42.548239</td>\n",
       "      <td>0.153888</td>\n",
       "      <td>0.171853</td>\n",
       "      <td>0.169448</td>\n",
       "      <td>0.170900</td>\n",
       "      <td>365.928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>3910.006923</td>\n",
       "      <td>5670.004885</td>\n",
       "      <td>1759.997962</td>\n",
       "      <td>0.125500</td>\n",
       "      <td>0.157136</td>\n",
       "      <td>3229.949353</td>\n",
       "      <td>42.548239</td>\n",
       "      <td>0.155718</td>\n",
       "      <td>0.176166</td>\n",
       "      <td>0.172699</td>\n",
       "      <td>0.173425</td>\n",
       "      <td>365.928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>3910.006923</td>\n",
       "      <td>5780.004758</td>\n",
       "      <td>1869.997835</td>\n",
       "      <td>0.125500</td>\n",
       "      <td>0.159149</td>\n",
       "      <td>3229.949353</td>\n",
       "      <td>42.548239</td>\n",
       "      <td>0.157362</td>\n",
       "      <td>0.180485</td>\n",
       "      <td>0.175838</td>\n",
       "      <td>0.175982</td>\n",
       "      <td>365.928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>3910.006923</td>\n",
       "      <td>5890.004630</td>\n",
       "      <td>1979.997707</td>\n",
       "      <td>0.125500</td>\n",
       "      <td>0.161165</td>\n",
       "      <td>3229.949353</td>\n",
       "      <td>42.548239</td>\n",
       "      <td>0.159181</td>\n",
       "      <td>0.184898</td>\n",
       "      <td>0.178798</td>\n",
       "      <td>0.178503</td>\n",
       "      <td>365.928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>3910.006923</td>\n",
       "      <td>6000.004503</td>\n",
       "      <td>2089.997580</td>\n",
       "      <td>0.125500</td>\n",
       "      <td>0.163183</td>\n",
       "      <td>3229.949353</td>\n",
       "      <td>42.548239</td>\n",
       "      <td>0.161044</td>\n",
       "      <td>0.189130</td>\n",
       "      <td>0.181635</td>\n",
       "      <td>0.180668</td>\n",
       "      <td>365.928</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2300 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pb            P           ΔP       μob         μ         Rsob  \\\n",
       "0   2200.008903  2200.008903     0.000000  0.998677  0.998677   478.607857   \n",
       "1   2200.008903  2348.166747   148.157844  0.998677  1.007916   478.607857   \n",
       "2   2200.008903  2496.324590   296.315687  0.998677  1.017489   478.607857   \n",
       "3   2200.008903  2644.482434   444.473531  0.998677  1.027342   478.607857   \n",
       "4   2200.008903  2792.640278   592.631375  0.998677  1.037429   478.607857   \n",
       "..          ...          ...          ...       ...       ...          ...   \n",
       "15  3910.006923  5560.005012  1649.998089  0.125500  0.155127  3229.949353   \n",
       "16  3910.006923  5670.004885  1759.997962  0.125500  0.157136  3229.949353   \n",
       "17  3910.006923  5780.004758  1869.997835  0.125500  0.159149  3229.949353   \n",
       "18  3910.006923  5890.004630  1979.997707  0.125500  0.161165  3229.949353   \n",
       "19  3910.006923  6000.004503  2089.997580  0.125500  0.163183  3229.949353   \n",
       "\n",
       "          API  sequential  functional  3layerdeep      poly       Tr  \n",
       "0   32.545608    0.998470    0.998585    0.998694  0.998792  336.150  \n",
       "1   32.545608    1.007414    1.007508    1.008885  1.008426  336.150  \n",
       "2   32.545608    1.015576    1.018905    1.015306  1.018305  336.150  \n",
       "3   32.545608    1.027691    1.027661    1.020632  1.033516  336.150  \n",
       "4   32.545608    1.039029    1.038241    1.027324  1.039544  336.150  \n",
       "..        ...         ...         ...         ...       ...      ...  \n",
       "15  42.548239    0.153888    0.171853    0.169448  0.170900  365.928  \n",
       "16  42.548239    0.155718    0.176166    0.172699  0.173425  365.928  \n",
       "17  42.548239    0.157362    0.180485    0.175838  0.175982  365.928  \n",
       "18  42.548239    0.159181    0.184898    0.178798  0.178503  365.928  \n",
       "19  42.548239    0.161044    0.189130    0.181635  0.180668  365.928  \n",
       "\n",
       "[2300 rows x 12 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "totaldf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1d2b7225",
   "metadata": {},
   "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>% AE</th>\n",
       "      <th>% Std Dev AE</th>\n",
       "      <th>%AAE</th>\n",
       "      <th>% Std Dev AAE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 hidden layer</th>\n",
       "      <td>-0.801810</td>\n",
       "      <td>4.792236</td>\n",
       "      <td>2.784125</td>\n",
       "      <td>3.982095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 hidden layers</th>\n",
       "      <td>-1.208490</td>\n",
       "      <td>5.751127</td>\n",
       "      <td>3.344563</td>\n",
       "      <td>4.832164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 hidden layers</th>\n",
       "      <td>-0.079283</td>\n",
       "      <td>7.010739</td>\n",
       "      <td>3.608539</td>\n",
       "      <td>6.011255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 hidden layers</th>\n",
       "      <td>1.319429</td>\n",
       "      <td>7.498708</td>\n",
       "      <td>3.678392</td>\n",
       "      <td>6.666404</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     % AE  % Std Dev AE      %AAE  % Std Dev AAE\n",
       "1 hidden layer  -0.801810      4.792236  2.784125       3.982095\n",
       "2 hidden layers -1.208490      5.751127  3.344563       4.832164\n",
       "3 hidden layers -0.079283      7.010739  3.608539       6.011255\n",
       "4 hidden layers  1.319429      7.498708  3.678392       6.666404"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def Stats_matrix(a):\n",
    "    average_error = []\n",
    "    standard_deviation = []\n",
    "    absolute_average_error = []\n",
    "    absolute_standard_deviation =[]\n",
    "    for i in range(4):\n",
    "        average_error.append(np.average((a.iloc[:,i+7]-a.μ)/a.μ*100))\n",
    "        standard_deviation.append(np.std((a.iloc[:,i+7]-a.μ)/a.μ*100))\n",
    "        absolute_average_error.append(np.average(np.absolute((a.iloc[:,i+7]-a.μ)/a.μ*100)))\n",
    "        absolute_standard_deviation.append(np.std(np.absolute((a.iloc[:,i+7]-a.μ)/totaldf.μ*100)))\n",
    "    statistics  = pd.DataFrame(list(zip(average_error,standard_deviation,absolute_average_error,\n",
    "                                        absolute_standard_deviation)),columns = ['% AE','% Std Dev AE','%AAE','% Std Dev AAE'],index = ['1 hidden layer','2 hidden layers','3 hidden layers','4 hidden layers'])\n",
    "    return statistics\n",
    "\n",
    "Stats_matrix(totaldf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f102cfdd",
   "metadata": {},
   "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>% AE</th>\n",
       "      <th>% Std Dev AE</th>\n",
       "      <th>%AAE</th>\n",
       "      <th>% Std Dev AAE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 hidden layer</th>\n",
       "      <td>-0.030773</td>\n",
       "      <td>3.539768</td>\n",
       "      <td>2.159435</td>\n",
       "      <td>3.486718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 hidden layers</th>\n",
       "      <td>-0.969527</td>\n",
       "      <td>3.921185</td>\n",
       "      <td>2.686268</td>\n",
       "      <td>5.242559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 hidden layers</th>\n",
       "      <td>-0.215188</td>\n",
       "      <td>3.864919</td>\n",
       "      <td>2.441804</td>\n",
       "      <td>5.365111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 hidden layers</th>\n",
       "      <td>1.458672</td>\n",
       "      <td>4.850759</td>\n",
       "      <td>2.817161</td>\n",
       "      <td>5.564974</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     % AE  % Std Dev AE      %AAE  % Std Dev AAE\n",
       "1 hidden layer  -0.030773      3.539768  2.159435       3.486718\n",
       "2 hidden layers -0.969527      3.921185  2.686268       5.242559\n",
       "3 hidden layers -0.215188      3.864919  2.441804       5.365111\n",
       "4 hidden layers  1.458672      4.850759  2.817161       5.564974"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visc_0_1 = totaldf.loc[(totaldf.μob<1) & (totaldf.μob>0)]\n",
    "Stats_matrix(visc_0_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a3d1302a",
   "metadata": {},
   "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>% AE</th>\n",
       "      <th>% Std Dev AE</th>\n",
       "      <th>%AAE</th>\n",
       "      <th>% Std Dev AAE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 hidden layer</th>\n",
       "      <td>-2.965636</td>\n",
       "      <td>6.330885</td>\n",
       "      <td>4.259406</td>\n",
       "      <td>96.989282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 hidden layers</th>\n",
       "      <td>-3.686089</td>\n",
       "      <td>6.149596</td>\n",
       "      <td>4.191746</td>\n",
       "      <td>100.393868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 hidden layers</th>\n",
       "      <td>-2.774424</td>\n",
       "      <td>6.446235</td>\n",
       "      <td>4.233429</td>\n",
       "      <td>101.463459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 hidden layers</th>\n",
       "      <td>-1.539563</td>\n",
       "      <td>5.870398</td>\n",
       "      <td>3.629105</td>\n",
       "      <td>88.478391</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     % AE  % Std Dev AE      %AAE  % Std Dev AAE\n",
       "1 hidden layer  -2.965636      6.330885  4.259406      96.989282\n",
       "2 hidden layers -3.686089      6.149596  4.191746     100.393868\n",
       "3 hidden layers -2.774424      6.446235  4.233429     101.463459\n",
       "4 hidden layers -1.539563      5.870398  3.629105      88.478391"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visc_1_5 = totaldf.loc[(totaldf.μob>1) & (totaldf.μob<5)]\n",
    "Stats_matrix(visc_1_5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4805b22f",
   "metadata": {},
   "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>% AE</th>\n",
       "      <th>% Std Dev AE</th>\n",
       "      <th>%AAE</th>\n",
       "      <th>% Std Dev AAE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 hidden layer</th>\n",
       "      <td>-1.665715</td>\n",
       "      <td>4.410642</td>\n",
       "      <td>2.657362</td>\n",
       "      <td>127.086197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 hidden layers</th>\n",
       "      <td>-1.540274</td>\n",
       "      <td>4.419241</td>\n",
       "      <td>2.593524</td>\n",
       "      <td>159.086237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 hidden layers</th>\n",
       "      <td>-0.123858</td>\n",
       "      <td>5.486818</td>\n",
       "      <td>3.560974</td>\n",
       "      <td>208.369249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 hidden layers</th>\n",
       "      <td>1.092513</td>\n",
       "      <td>5.426765</td>\n",
       "      <td>3.796174</td>\n",
       "      <td>197.235671</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     % AE  % Std Dev AE      %AAE  % Std Dev AAE\n",
       "1 hidden layer  -1.665715      4.410642  2.657362     127.086197\n",
       "2 hidden layers -1.540274      4.419241  2.593524     159.086237\n",
       "3 hidden layers -0.123858      5.486818  3.560974     208.369249\n",
       "4 hidden layers  1.092513      5.426765  3.796174     197.235671"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visc_5_20 = totaldf.loc[(totaldf.μob>5) & (totaldf.μob<20)]\n",
    "Stats_matrix(visc_5_20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3cb7e477",
   "metadata": {},
   "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>% AE</th>\n",
       "      <th>% Std Dev AE</th>\n",
       "      <th>%AAE</th>\n",
       "      <th>% Std Dev AAE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1 hidden layer</th>\n",
       "      <td>2.007498</td>\n",
       "      <td>6.614478</td>\n",
       "      <td>3.992433</td>\n",
       "      <td>1163.478911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2 hidden layers</th>\n",
       "      <td>9.573543</td>\n",
       "      <td>11.867589</td>\n",
       "      <td>9.920049</td>\n",
       "      <td>2257.443042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3 hidden layers</th>\n",
       "      <td>16.511302</td>\n",
       "      <td>17.520383</td>\n",
       "      <td>16.915920</td>\n",
       "      <td>3761.729543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4 hidden layers</th>\n",
       "      <td>15.325332</td>\n",
       "      <td>21.487327</td>\n",
       "      <td>15.891347</td>\n",
       "      <td>4272.999364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      % AE  % Std Dev AE       %AAE  % Std Dev AAE\n",
       "1 hidden layer    2.007498      6.614478   3.992433    1163.478911\n",
       "2 hidden layers   9.573543     11.867589   9.920049    2257.443042\n",
       "3 hidden layers  16.511302     17.520383  16.915920    3761.729543\n",
       "4 hidden layers  15.325332     21.487327  15.891347    4272.999364"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visc_20_plus = totaldf.loc[(totaldf.μob>20) ]\n",
    "Stats_matrix(visc_20_plus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0607533b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, '% Average Absolute Error')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "press_0_25 = totaldf.loc[(totaldf.ΔP<2500)]\n",
    "press_25_50 = totaldf.loc[(totaldf.ΔP<5000)&(totaldf.ΔP>2500)]\n",
    "press_50_100 = totaldf.loc[(totaldf.ΔP<10000)&(totaldf.ΔP>5000)]\n",
    "visc_0_1 = totaldf.loc[(totaldf.μob>0) & (totaldf.μob<1)]\n",
    "visc_1_5 = totaldf.loc[(totaldf.μob>1) & (totaldf.μob<5)]\n",
    "visc_5_20 = totaldf.loc[(totaldf.μob>5) & (totaldf.μob<20)]\n",
    "visc_20_plus = totaldf.loc[(totaldf.μob>20) ]\n",
    "\n",
    "full_data = {\n",
    "    'Pdif 0-2500' : Stats_matrix(press_0_25).iloc[:,2],\n",
    "    'Pdif 2500-5000' : Stats_matrix(press_25_50).iloc[:,2],\n",
    "    'Pdif 5000-10000' : Stats_matrix(press_50_100).iloc[:,2],\n",
    "    'BP V 0-1' : Stats_matrix(visc_0_1).iloc[:,2],\n",
    "    'BP V 1-5' : Stats_matrix(visc_1_5).iloc[:,2],\n",
    "    'BP V 5-20' : Stats_matrix(visc_5_20).iloc[:,2],\n",
    "    'BP V 20+' : Stats_matrix(visc_20_plus).iloc[:,2],\n",
    "}   \n",
    "models =   ['1 hidden layer','2 hidden layers','3 hidden layers','4 hidden layers']\n",
    "\n",
    "%run my_bar_plot.ipynb\n",
    "fig, ax = plt.subplots()\n",
    "plt.rcParams['figure.figsize'] = (15, 8)\n",
    "bar_plot(ax, full_data, total_width=.8, single_width=.9,)\n",
    "plt.xticks(range(4), models)\n",
    "plt.title('Undersaturated Oil Viscosity ')\n",
    "plt.ylabel(\"% Average Absolute Error\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2f73c069",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": 5
}