Codon_usage_Project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split as tts\n",
    "# from sklearn.preprocessing import minmax_scale\n",
    "\n",
    "from scipy.stats import norm, gaussian_kde\n",
    "\n",
    "from imblearn.over_sampling import RandomOverSampler, SMOTE, ADASYN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Download the date\n",
    "https://archive.ics.uci.edu/ml/machine-learning-databases/00577/codon_usage.csv.zip\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# !wget https://archive.ics.uci.edu/ml/machine-learning-databases/00577/codon_usage.csv.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# !unzip codon_usage.csv.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def read_data(file_name):\n",
    "    \"\"\"\n",
    "    Reads in a csv file and returns a dataframe\n",
    "    \"\"\"\n",
    "    return pd.read_csv(file_name, low_memory=False)\n",
    "\n",
    "dataset = read_data('codon_usage.csv')\n",
    "df = dataset.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>DNAtype</th>\n",
       "      <th>SpeciesID</th>\n",
       "      <th>Ncodons</th>\n",
       "      <th>UUA</th>\n",
       "      <th>UUG</th>\n",
       "      <th>CUU</th>\n",
       "      <th>CUC</th>\n",
       "      <th>CUA</th>\n",
       "      <th>CUG</th>\n",
       "      <th>AUU</th>\n",
       "      <th>...</th>\n",
       "      <th>CGG</th>\n",
       "      <th>AGA</th>\n",
       "      <th>AGG</th>\n",
       "      <th>GAU</th>\n",
       "      <th>GAC</th>\n",
       "      <th>GAA</th>\n",
       "      <th>GAG</th>\n",
       "      <th>UAA</th>\n",
       "      <th>UAG</th>\n",
       "      <th>UGA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>1.302800e+04</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "      <td>13028.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.367209</td>\n",
       "      <td>130451.105926</td>\n",
       "      <td>7.960576e+04</td>\n",
       "      <td>0.020637</td>\n",
       "      <td>0.014104</td>\n",
       "      <td>0.017820</td>\n",
       "      <td>0.018288</td>\n",
       "      <td>0.019044</td>\n",
       "      <td>0.018450</td>\n",
       "      <td>0.028352</td>\n",
       "      <td>...</td>\n",
       "      <td>0.005454</td>\n",
       "      <td>0.009929</td>\n",
       "      <td>0.006422</td>\n",
       "      <td>0.024178</td>\n",
       "      <td>0.021164</td>\n",
       "      <td>0.028290</td>\n",
       "      <td>0.021683</td>\n",
       "      <td>0.001645</td>\n",
       "      <td>0.000592</td>\n",
       "      <td>0.006178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.688726</td>\n",
       "      <td>124787.086107</td>\n",
       "      <td>7.197010e+05</td>\n",
       "      <td>0.020709</td>\n",
       "      <td>0.009280</td>\n",
       "      <td>0.010586</td>\n",
       "      <td>0.014572</td>\n",
       "      <td>0.024250</td>\n",
       "      <td>0.016578</td>\n",
       "      <td>0.017507</td>\n",
       "      <td>...</td>\n",
       "      <td>0.006605</td>\n",
       "      <td>0.008574</td>\n",
       "      <td>0.006387</td>\n",
       "      <td>0.013828</td>\n",
       "      <td>0.013041</td>\n",
       "      <td>0.014342</td>\n",
       "      <td>0.015018</td>\n",
       "      <td>0.001834</td>\n",
       "      <td>0.000907</td>\n",
       "      <td>0.010344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>1.000000e+03</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>28850.750000</td>\n",
       "      <td>1.602000e+03</td>\n",
       "      <td>0.005610</td>\n",
       "      <td>0.007108</td>\n",
       "      <td>0.010890</td>\n",
       "      <td>0.007830</td>\n",
       "      <td>0.005307</td>\n",
       "      <td>0.007180</td>\n",
       "      <td>0.016360</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001220</td>\n",
       "      <td>0.001690</td>\n",
       "      <td>0.001170</td>\n",
       "      <td>0.012380</td>\n",
       "      <td>0.011860</td>\n",
       "      <td>0.017360</td>\n",
       "      <td>0.009710</td>\n",
       "      <td>0.000560</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>81971.500000</td>\n",
       "      <td>2.927500e+03</td>\n",
       "      <td>0.015260</td>\n",
       "      <td>0.013360</td>\n",
       "      <td>0.016130</td>\n",
       "      <td>0.014560</td>\n",
       "      <td>0.009685</td>\n",
       "      <td>0.012800</td>\n",
       "      <td>0.025475</td>\n",
       "      <td>...</td>\n",
       "      <td>0.003530</td>\n",
       "      <td>0.009270</td>\n",
       "      <td>0.004545</td>\n",
       "      <td>0.025420</td>\n",
       "      <td>0.019070</td>\n",
       "      <td>0.026085</td>\n",
       "      <td>0.020540</td>\n",
       "      <td>0.001380</td>\n",
       "      <td>0.000420</td>\n",
       "      <td>0.001130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>222891.250000</td>\n",
       "      <td>9.120000e+03</td>\n",
       "      <td>0.029485</td>\n",
       "      <td>0.019810</td>\n",
       "      <td>0.022730</td>\n",
       "      <td>0.025112</td>\n",
       "      <td>0.017245</td>\n",
       "      <td>0.024315</td>\n",
       "      <td>0.038113</td>\n",
       "      <td>...</td>\n",
       "      <td>0.007150</td>\n",
       "      <td>0.015922</td>\n",
       "      <td>0.010250</td>\n",
       "      <td>0.034190</td>\n",
       "      <td>0.027690</td>\n",
       "      <td>0.036800</td>\n",
       "      <td>0.031122</td>\n",
       "      <td>0.002370</td>\n",
       "      <td>0.000830</td>\n",
       "      <td>0.002890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>12.000000</td>\n",
       "      <td>465364.000000</td>\n",
       "      <td>4.066258e+07</td>\n",
       "      <td>0.151330</td>\n",
       "      <td>0.101190</td>\n",
       "      <td>0.089780</td>\n",
       "      <td>0.100350</td>\n",
       "      <td>0.163920</td>\n",
       "      <td>0.107370</td>\n",
       "      <td>0.154060</td>\n",
       "      <td>...</td>\n",
       "      <td>0.055540</td>\n",
       "      <td>0.098830</td>\n",
       "      <td>0.058430</td>\n",
       "      <td>0.185660</td>\n",
       "      <td>0.113840</td>\n",
       "      <td>0.144890</td>\n",
       "      <td>0.158550</td>\n",
       "      <td>0.045200</td>\n",
       "      <td>0.025610</td>\n",
       "      <td>0.106700</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            DNAtype      SpeciesID       Ncodons           UUA           UUG  \\\n",
       "count  13028.000000   13028.000000  1.302800e+04  13028.000000  13028.000000   \n",
       "mean       0.367209  130451.105926  7.960576e+04      0.020637      0.014104   \n",
       "std        0.688726  124787.086107  7.197010e+05      0.020709      0.009280   \n",
       "min        0.000000       7.000000  1.000000e+03      0.000000      0.000000   \n",
       "25%        0.000000   28850.750000  1.602000e+03      0.005610      0.007108   \n",
       "50%        0.000000   81971.500000  2.927500e+03      0.015260      0.013360   \n",
       "75%        1.000000  222891.250000  9.120000e+03      0.029485      0.019810   \n",
       "max       12.000000  465364.000000  4.066258e+07      0.151330      0.101190   \n",
       "\n",
       "                CUU           CUC           CUA           CUG           AUU  \\\n",
       "count  13028.000000  13028.000000  13028.000000  13028.000000  13028.000000   \n",
       "mean       0.017820      0.018288      0.019044      0.018450      0.028352   \n",
       "std        0.010586      0.014572      0.024250      0.016578      0.017507   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.010890      0.007830      0.005307      0.007180      0.016360   \n",
       "50%        0.016130      0.014560      0.009685      0.012800      0.025475   \n",
       "75%        0.022730      0.025112      0.017245      0.024315      0.038113   \n",
       "max        0.089780      0.100350      0.163920      0.107370      0.154060   \n",
       "\n",
       "       ...           CGG           AGA           AGG           GAU  \\\n",
       "count  ...  13028.000000  13028.000000  13028.000000  13028.000000   \n",
       "mean   ...      0.005454      0.009929      0.006422      0.024178   \n",
       "std    ...      0.006605      0.008574      0.006387      0.013828   \n",
       "min    ...      0.000000      0.000000      0.000000      0.000000   \n",
       "25%    ...      0.001220      0.001690      0.001170      0.012380   \n",
       "50%    ...      0.003530      0.009270      0.004545      0.025420   \n",
       "75%    ...      0.007150      0.015922      0.010250      0.034190   \n",
       "max    ...      0.055540      0.098830      0.058430      0.185660   \n",
       "\n",
       "                GAC           GAA           GAG           UAA           UAG  \\\n",
       "count  13028.000000  13028.000000  13028.000000  13028.000000  13028.000000   \n",
       "mean       0.021164      0.028290      0.021683      0.001645      0.000592   \n",
       "std        0.013041      0.014342      0.015018      0.001834      0.000907   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.011860      0.017360      0.009710      0.000560      0.000000   \n",
       "50%        0.019070      0.026085      0.020540      0.001380      0.000420   \n",
       "75%        0.027690      0.036800      0.031122      0.002370      0.000830   \n",
       "max        0.113840      0.144890      0.158550      0.045200      0.025610   \n",
       "\n",
       "                UGA  \n",
       "count  13028.000000  \n",
       "mean       0.006178  \n",
       "std        0.010344  \n",
       "min        0.000000  \n",
       "25%        0.000410  \n",
       "50%        0.001130  \n",
       "75%        0.002890  \n",
       "max        0.106700  \n",
       "\n",
       "[8 rows x 65 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dataset = read_data('codon_usage.csv')\n",
    "dataset.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>Kingdom</th>\n",
       "      <th>DNAtype</th>\n",
       "      <th>SpeciesID</th>\n",
       "      <th>Ncodons</th>\n",
       "      <th>SpeciesName</th>\n",
       "      <th>UUU</th>\n",
       "      <th>UUC</th>\n",
       "      <th>UUA</th>\n",
       "      <th>UUG</th>\n",
       "      <th>CUU</th>\n",
       "      <th>...</th>\n",
       "      <th>CGG</th>\n",
       "      <th>AGA</th>\n",
       "      <th>AGG</th>\n",
       "      <th>GAU</th>\n",
       "      <th>GAC</th>\n",
       "      <th>GAA</th>\n",
       "      <th>GAG</th>\n",
       "      <th>UAA</th>\n",
       "      <th>UAG</th>\n",
       "      <th>UGA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>vrl</td>\n",
       "      <td>0</td>\n",
       "      <td>100217</td>\n",
       "      <td>1995</td>\n",
       "      <td>Epizootic haematopoietic necrosis virus</td>\n",
       "      <td>0.01654</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>0.00050</td>\n",
       "      <td>0.00351</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00451</td>\n",
       "      <td>0.01303</td>\n",
       "      <td>0.03559</td>\n",
       "      <td>0.01003</td>\n",
       "      <td>0.04612</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>0.04361</td>\n",
       "      <td>0.00251</td>\n",
       "      <td>0.00050</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>vrl</td>\n",
       "      <td>0</td>\n",
       "      <td>100220</td>\n",
       "      <td>1474</td>\n",
       "      <td>Bohle iridovirus</td>\n",
       "      <td>0.02714</td>\n",
       "      <td>0.01357</td>\n",
       "      <td>0.00068</td>\n",
       "      <td>0.00678</td>\n",
       "      <td>0.00407</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00136</td>\n",
       "      <td>0.01696</td>\n",
       "      <td>0.03596</td>\n",
       "      <td>0.01221</td>\n",
       "      <td>0.04545</td>\n",
       "      <td>0.01560</td>\n",
       "      <td>0.04410</td>\n",
       "      <td>0.00271</td>\n",
       "      <td>0.00068</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>vrl</td>\n",
       "      <td>0</td>\n",
       "      <td>100755</td>\n",
       "      <td>4862</td>\n",
       "      <td>Sweet potato leaf curl virus</td>\n",
       "      <td>0.01974</td>\n",
       "      <td>0.0218</td>\n",
       "      <td>0.01357</td>\n",
       "      <td>0.01543</td>\n",
       "      <td>0.00782</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00596</td>\n",
       "      <td>0.01974</td>\n",
       "      <td>0.02489</td>\n",
       "      <td>0.03126</td>\n",
       "      <td>0.02036</td>\n",
       "      <td>0.02242</td>\n",
       "      <td>0.02468</td>\n",
       "      <td>0.00391</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>vrl</td>\n",
       "      <td>0</td>\n",
       "      <td>100880</td>\n",
       "      <td>1915</td>\n",
       "      <td>Northern cereal mosaic virus</td>\n",
       "      <td>0.01775</td>\n",
       "      <td>0.02245</td>\n",
       "      <td>0.01619</td>\n",
       "      <td>0.00992</td>\n",
       "      <td>0.01567</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00366</td>\n",
       "      <td>0.01410</td>\n",
       "      <td>0.01671</td>\n",
       "      <td>0.03760</td>\n",
       "      <td>0.01932</td>\n",
       "      <td>0.03029</td>\n",
       "      <td>0.03446</td>\n",
       "      <td>0.00261</td>\n",
       "      <td>0.00157</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>vrl</td>\n",
       "      <td>0</td>\n",
       "      <td>100887</td>\n",
       "      <td>22831</td>\n",
       "      <td>Soil-borne cereal mosaic virus</td>\n",
       "      <td>0.02816</td>\n",
       "      <td>0.01371</td>\n",
       "      <td>0.00767</td>\n",
       "      <td>0.03679</td>\n",
       "      <td>0.01380</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00604</td>\n",
       "      <td>0.01494</td>\n",
       "      <td>0.01734</td>\n",
       "      <td>0.04148</td>\n",
       "      <td>0.02483</td>\n",
       "      <td>0.03359</td>\n",
       "      <td>0.03679</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00044</td>\n",
       "      <td>0.00131</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>13023</th>\n",
       "      <td>pri</td>\n",
       "      <td>0</td>\n",
       "      <td>9601</td>\n",
       "      <td>1097</td>\n",
       "      <td>Pongo pygmaeus abelii</td>\n",
       "      <td>0.02552</td>\n",
       "      <td>0.03555</td>\n",
       "      <td>0.00547</td>\n",
       "      <td>0.01367</td>\n",
       "      <td>0.01276</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00820</td>\n",
       "      <td>0.01367</td>\n",
       "      <td>0.01094</td>\n",
       "      <td>0.01367</td>\n",
       "      <td>0.02279</td>\n",
       "      <td>0.02005</td>\n",
       "      <td>0.04102</td>\n",
       "      <td>0.00091</td>\n",
       "      <td>0.00091</td>\n",
       "      <td>0.00638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13024</th>\n",
       "      <td>pri</td>\n",
       "      <td>1</td>\n",
       "      <td>9601</td>\n",
       "      <td>2067</td>\n",
       "      <td>mitochondrion Pongo pygmaeus abelii</td>\n",
       "      <td>0.01258</td>\n",
       "      <td>0.03193</td>\n",
       "      <td>0.01984</td>\n",
       "      <td>0.00629</td>\n",
       "      <td>0.01451</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00145</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00048</td>\n",
       "      <td>0.00194</td>\n",
       "      <td>0.01306</td>\n",
       "      <td>0.01838</td>\n",
       "      <td>0.00677</td>\n",
       "      <td>0.00242</td>\n",
       "      <td>0.00097</td>\n",
       "      <td>0.01887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13025</th>\n",
       "      <td>pri</td>\n",
       "      <td>1</td>\n",
       "      <td>9602</td>\n",
       "      <td>1686</td>\n",
       "      <td>mitochondrion Pongo pygmaeus pygmaeus</td>\n",
       "      <td>0.01423</td>\n",
       "      <td>0.03321</td>\n",
       "      <td>0.01661</td>\n",
       "      <td>0.00356</td>\n",
       "      <td>0.01127</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00178</td>\n",
       "      <td>0.01661</td>\n",
       "      <td>0.02788</td>\n",
       "      <td>0.00297</td>\n",
       "      <td>0.00356</td>\n",
       "      <td>0.00119</td>\n",
       "      <td>0.02017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13026</th>\n",
       "      <td>pri</td>\n",
       "      <td>0</td>\n",
       "      <td>9606</td>\n",
       "      <td>40662582</td>\n",
       "      <td>Homo sapiens</td>\n",
       "      <td>0.01757</td>\n",
       "      <td>0.02028</td>\n",
       "      <td>0.00767</td>\n",
       "      <td>0.01293</td>\n",
       "      <td>0.01319</td>\n",
       "      <td>...</td>\n",
       "      <td>0.01142</td>\n",
       "      <td>0.01217</td>\n",
       "      <td>0.01196</td>\n",
       "      <td>0.02178</td>\n",
       "      <td>0.02510</td>\n",
       "      <td>0.02896</td>\n",
       "      <td>0.03959</td>\n",
       "      <td>0.00099</td>\n",
       "      <td>0.00079</td>\n",
       "      <td>0.00156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13027</th>\n",
       "      <td>pri</td>\n",
       "      <td>1</td>\n",
       "      <td>9606</td>\n",
       "      <td>8998998</td>\n",
       "      <td>mitochondrion Homo sapiens</td>\n",
       "      <td>0.01778</td>\n",
       "      <td>0.03724</td>\n",
       "      <td>0.01732</td>\n",
       "      <td>0.00600</td>\n",
       "      <td>0.01689</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00083</td>\n",
       "      <td>0.00041</td>\n",
       "      <td>0.00041</td>\n",
       "      <td>0.00451</td>\n",
       "      <td>0.01402</td>\n",
       "      <td>0.01651</td>\n",
       "      <td>0.00783</td>\n",
       "      <td>0.00156</td>\n",
       "      <td>0.00114</td>\n",
       "      <td>0.02161</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>13028 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Kingdom  DNAtype  SpeciesID   Ncodons  \\\n",
       "0         vrl        0     100217      1995   \n",
       "1         vrl        0     100220      1474   \n",
       "2         vrl        0     100755      4862   \n",
       "3         vrl        0     100880      1915   \n",
       "4         vrl        0     100887     22831   \n",
       "...       ...      ...        ...       ...   \n",
       "13023     pri        0       9601      1097   \n",
       "13024     pri        1       9601      2067   \n",
       "13025     pri        1       9602      1686   \n",
       "13026     pri        0       9606  40662582   \n",
       "13027     pri        1       9606   8998998   \n",
       "\n",
       "                                   SpeciesName      UUU      UUC      UUA  \\\n",
       "0      Epizootic haematopoietic necrosis virus  0.01654  0.01203  0.00050   \n",
       "1                             Bohle iridovirus  0.02714  0.01357  0.00068   \n",
       "2                 Sweet potato leaf curl virus  0.01974   0.0218  0.01357   \n",
       "3                 Northern cereal mosaic virus  0.01775  0.02245  0.01619   \n",
       "4               Soil-borne cereal mosaic virus  0.02816  0.01371  0.00767   \n",
       "...                                        ...      ...      ...      ...   \n",
       "13023                    Pongo pygmaeus abelii  0.02552  0.03555  0.00547   \n",
       "13024      mitochondrion Pongo pygmaeus abelii  0.01258  0.03193  0.01984   \n",
       "13025    mitochondrion Pongo pygmaeus pygmaeus  0.01423  0.03321  0.01661   \n",
       "13026                             Homo sapiens  0.01757  0.02028  0.00767   \n",
       "13027               mitochondrion Homo sapiens  0.01778  0.03724  0.01732   \n",
       "\n",
       "           UUG      CUU  ...      CGG      AGA      AGG      GAU      GAC  \\\n",
       "0      0.00351  0.01203  ...  0.00451  0.01303  0.03559  0.01003  0.04612   \n",
       "1      0.00678  0.00407  ...  0.00136  0.01696  0.03596  0.01221  0.04545   \n",
       "2      0.01543  0.00782  ...  0.00596  0.01974  0.02489  0.03126  0.02036   \n",
       "3      0.00992  0.01567  ...  0.00366  0.01410  0.01671  0.03760  0.01932   \n",
       "4      0.03679  0.01380  ...  0.00604  0.01494  0.01734  0.04148  0.02483   \n",
       "...        ...      ...  ...      ...      ...      ...      ...      ...   \n",
       "13023  0.01367  0.01276  ...  0.00820  0.01367  0.01094  0.01367  0.02279   \n",
       "13024  0.00629  0.01451  ...  0.00145  0.00000  0.00048  0.00194  0.01306   \n",
       "13025  0.00356  0.01127  ...  0.00000  0.00000  0.00000  0.00178  0.01661   \n",
       "13026  0.01293  0.01319  ...  0.01142  0.01217  0.01196  0.02178  0.02510   \n",
       "13027  0.00600  0.01689  ...  0.00083  0.00041  0.00041  0.00451  0.01402   \n",
       "\n",
       "           GAA      GAG      UAA      UAG      UGA  \n",
       "0      0.01203  0.04361  0.00251  0.00050  0.00000  \n",
       "1      0.01560  0.04410  0.00271  0.00068  0.00000  \n",
       "2      0.02242  0.02468  0.00391  0.00000  0.00144  \n",
       "3      0.03029  0.03446  0.00261  0.00157  0.00000  \n",
       "4      0.03359  0.03679  0.00000  0.00044  0.00131  \n",
       "...        ...      ...      ...      ...      ...  \n",
       "13023  0.02005  0.04102  0.00091  0.00091  0.00638  \n",
       "13024  0.01838  0.00677  0.00242  0.00097  0.01887  \n",
       "13025  0.02788  0.00297  0.00356  0.00119  0.02017  \n",
       "13026  0.02896  0.03959  0.00099  0.00079  0.00156  \n",
       "13027  0.01651  0.00783  0.00156  0.00114  0.02161  \n",
       "\n",
       "[13028 rows x 69 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bct    2920\n",
       "vrl    2832\n",
       "pln    2523\n",
       "vrt    2077\n",
       "inv    1345\n",
       "mam     572\n",
       "phg     220\n",
       "rod     215\n",
       "pri     180\n",
       "arc     126\n",
       "plm      18\n",
       "Name: Kingdom, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dataset.dropna(inplace=True)\n",
    "\n",
    "dataset['Kingdom'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Kingdom        0\n",
      "DNAtype        0\n",
      "SpeciesID      0\n",
      "Ncodons        0\n",
      "SpeciesName    0\n",
      "              ..\n",
      "GAA            0\n",
      "GAG            0\n",
      "UAA            0\n",
      "UAG            0\n",
      "UGA            0\n",
      "Length: 69, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(dataset.isnull().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset['Kingdom'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "classes_dict = {    'arc': 0, 'bct': 1, 'phg': 2, 'plm': 3, \n",
    "                    'pln': 4, 'inv': 5, 'vrt': 6, 'mam': 7,\n",
    "                    'rod': 8, 'pri': 9, 'vrl':10\n",
    "                }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "dataset['Kingdom'] = dataset['Kingdom'].map(classes_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "# dataset['Kingdom'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "dataset = dataset.drop(dataset[dataset['SpeciesID']==353569].index)\n",
    "dataset = dataset.drop(dataset[dataset['SpeciesID']==12440].index)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "dataset['UUU'] = dataset['UUU'].astype(float)\n",
    "dataset['UUC'] = dataset['UUC'].astype(float)\n",
    "# dataset = minmax_scale(dataset)\n",
    "# cols = dataset.select_dtypes(np.number).columns\n",
    "\n",
    "# num_columns_list = list(cols)\n",
    "# # num_columns_list\n",
    "\n",
    "# num_columns_list.remove('Kingdom')\n",
    "# # num_columns_list.remove('DNAtype')\n",
    "# num_columns_list.remove('SpeciesID')\n",
    "# # num_columns_list.remove('Ncodons')\n",
    "# # num_columns_list\n",
    "\n",
    "# Removing the minmax_scale of the columns because it is not necessary it might lead to 0\n",
    "# for num_column in num_columns_list:\n",
    "#     dataset[num_column] = minmax_scale(dataset[num_column]) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "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>Kingdom</th>\n",
       "      <th>DNAtype</th>\n",
       "      <th>SpeciesID</th>\n",
       "      <th>Ncodons</th>\n",
       "      <th>SpeciesName</th>\n",
       "      <th>UUU</th>\n",
       "      <th>UUC</th>\n",
       "      <th>UUA</th>\n",
       "      <th>UUG</th>\n",
       "      <th>CUU</th>\n",
       "      <th>...</th>\n",
       "      <th>CGG</th>\n",
       "      <th>AGA</th>\n",
       "      <th>AGG</th>\n",
       "      <th>GAU</th>\n",
       "      <th>GAC</th>\n",
       "      <th>GAA</th>\n",
       "      <th>GAG</th>\n",
       "      <th>UAA</th>\n",
       "      <th>UAG</th>\n",
       "      <th>UGA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>100217</td>\n",
       "      <td>1995</td>\n",
       "      <td>Epizootic haematopoietic necrosis virus</td>\n",
       "      <td>0.01654</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>0.00050</td>\n",
       "      <td>0.00351</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00451</td>\n",
       "      <td>0.01303</td>\n",
       "      <td>0.03559</td>\n",
       "      <td>0.01003</td>\n",
       "      <td>0.04612</td>\n",
       "      <td>0.01203</td>\n",
       "      <td>0.04361</td>\n",
       "      <td>0.00251</td>\n",
       "      <td>0.00050</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>100220</td>\n",
       "      <td>1474</td>\n",
       "      <td>Bohle iridovirus</td>\n",
       "      <td>0.02714</td>\n",
       "      <td>0.01357</td>\n",
       "      <td>0.00068</td>\n",
       "      <td>0.00678</td>\n",
       "      <td>0.00407</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00136</td>\n",
       "      <td>0.01696</td>\n",
       "      <td>0.03596</td>\n",
       "      <td>0.01221</td>\n",
       "      <td>0.04545</td>\n",
       "      <td>0.01560</td>\n",
       "      <td>0.04410</td>\n",
       "      <td>0.00271</td>\n",
       "      <td>0.00068</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>100755</td>\n",
       "      <td>4862</td>\n",
       "      <td>Sweet potato leaf curl virus</td>\n",
       "      <td>0.01974</td>\n",
       "      <td>0.02180</td>\n",
       "      <td>0.01357</td>\n",
       "      <td>0.01543</td>\n",
       "      <td>0.00782</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00596</td>\n",
       "      <td>0.01974</td>\n",
       "      <td>0.02489</td>\n",
       "      <td>0.03126</td>\n",
       "      <td>0.02036</td>\n",
       "      <td>0.02242</td>\n",
       "      <td>0.02468</td>\n",
       "      <td>0.00391</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>100880</td>\n",
       "      <td>1915</td>\n",
       "      <td>Northern cereal mosaic virus</td>\n",
       "      <td>0.01775</td>\n",
       "      <td>0.02245</td>\n",
       "      <td>0.01619</td>\n",
       "      <td>0.00992</td>\n",
       "      <td>0.01567</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00366</td>\n",
       "      <td>0.01410</td>\n",
       "      <td>0.01671</td>\n",
       "      <td>0.03760</td>\n",
       "      <td>0.01932</td>\n",
       "      <td>0.03029</td>\n",
       "      <td>0.03446</td>\n",
       "      <td>0.00261</td>\n",
       "      <td>0.00157</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>100887</td>\n",
       "      <td>22831</td>\n",
       "      <td>Soil-borne cereal mosaic virus</td>\n",
       "      <td>0.02816</td>\n",
       "      <td>0.01371</td>\n",
       "      <td>0.00767</td>\n",
       "      <td>0.03679</td>\n",
       "      <td>0.01380</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00604</td>\n",
       "      <td>0.01494</td>\n",
       "      <td>0.01734</td>\n",
       "      <td>0.04148</td>\n",
       "      <td>0.02483</td>\n",
       "      <td>0.03359</td>\n",
       "      <td>0.03679</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00044</td>\n",
       "      <td>0.00131</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Kingdom  DNAtype  SpeciesID  Ncodons  \\\n",
       "0       10        0     100217     1995   \n",
       "1       10        0     100220     1474   \n",
       "2       10        0     100755     4862   \n",
       "3       10        0     100880     1915   \n",
       "4       10        0     100887    22831   \n",
       "\n",
       "                               SpeciesName      UUU      UUC      UUA  \\\n",
       "0  Epizootic haematopoietic necrosis virus  0.01654  0.01203  0.00050   \n",
       "1                         Bohle iridovirus  0.02714  0.01357  0.00068   \n",
       "2             Sweet potato leaf curl virus  0.01974  0.02180  0.01357   \n",
       "3             Northern cereal mosaic virus  0.01775  0.02245  0.01619   \n",
       "4           Soil-borne cereal mosaic virus  0.02816  0.01371  0.00767   \n",
       "\n",
       "       UUG      CUU  ...      CGG      AGA      AGG      GAU      GAC  \\\n",
       "0  0.00351  0.01203  ...  0.00451  0.01303  0.03559  0.01003  0.04612   \n",
       "1  0.00678  0.00407  ...  0.00136  0.01696  0.03596  0.01221  0.04545   \n",
       "2  0.01543  0.00782  ...  0.00596  0.01974  0.02489  0.03126  0.02036   \n",
       "3  0.00992  0.01567  ...  0.00366  0.01410  0.01671  0.03760  0.01932   \n",
       "4  0.03679  0.01380  ...  0.00604  0.01494  0.01734  0.04148  0.02483   \n",
       "\n",
       "       GAA      GAG      UAA      UAG      UGA  \n",
       "0  0.01203  0.04361  0.00251  0.00050  0.00000  \n",
       "1  0.01560  0.04410  0.00271  0.00068  0.00000  \n",
       "2  0.02242  0.02468  0.00391  0.00000  0.00144  \n",
       "3  0.03029  0.03446  0.00261  0.00157  0.00000  \n",
       "4  0.03359  0.03679  0.00000  0.00044  0.00131  \n",
       "\n",
       "[5 rows x 69 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "X = dataset.iloc[:, 1:]\n",
    "X.drop(['SpeciesName', 'SpeciesID', 'DNAtype', 'Ncodons'], axis=1, inplace=True)\n",
    "X\n",
    "X_original = X.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "X = X.values.astype(float)\n",
    "y = dataset.iloc[:, 0]\n",
    "y_original = y.values.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "class DataGenerator:\n",
    "\n",
    "    def __init__(self, X, y, sampleSize=16):\n",
    "        self.X = X\n",
    "        self.y = y\n",
    "        self.sampleSize = sampleSize\n",
    "        if self.sampleSize <= 0 and self.sampleSize > len(self.X):\n",
    "            return TypeError(\"sampleSize must be between 0 and \", len(X))\n",
    "\n",
    "    def split_data(self, test_size=0.2, random_state=1):\n",
    "        self.X_train, self.X_rem, self.y_train, self.y_rem = tts(self.X, self.y, test_size=test_size, random_state=random_state)\n",
    "        return self.X_train, self.X_rem, self.y_train, self.y_rem\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13026, 64)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13026,)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "data_generator = DataGenerator(y=y, X=X)\n",
    "X_training, X_testing, y_training, y_testing = data_generator.split_data()\n",
    "X_training_original = X_training.copy()\n",
    "y_training_original = y_training.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def sampler(X, y, technique):\n",
    "    if technique =='ros':\n",
    "        ros = RandomOverSampler(random_state=1)\n",
    "        X_training, y_training = ros.fit_resample(X, y)\n",
    "    \n",
    "    elif technique =='smote':\n",
    "        smoothing = SMOTE(random_state=1)\n",
    "        X_training, y_training = smoothing.fit_resample(X, y)\n",
    "    \n",
    "    elif technique =='adasyn':\n",
    "        adasyn = ADASYN(random_state=1)\n",
    "        X_training, y_training = adasyn.fit_resample(X, y)\n",
    "    \n",
    "    return X_training, y_training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "class MultiClassClassification:\n",
    "    def __init__(self, X_train, y_train, laplace_smoothing_value, laplace, total_classes=11, for_arc = False, for_plm = False):\n",
    "        self.laplace_smoothing_value = laplace_smoothing_value\n",
    "        self.laplace = laplace\n",
    "        self.total_classes = total_classes\n",
    "        self.X_train = X_train\n",
    "        self.y_train = y_train\n",
    "        self.for_arc = for_arc\n",
    "        self.for_plm = for_plm\n",
    "        self.fitFeatures = []\n",
    "        self.list_of_ranges = []\n",
    "        self.set_of_ranges = set([])\n",
    "\n",
    "    def fit_distribution(self, data):\n",
    "\n",
    "        # mean = np.mean(data) # mean of the data\n",
    "        # std = np.std(data) # standard deviation of the data\n",
    "        # dist = norm(mean, std) # create a normal distribution with the mean and standard deviation\n",
    "\n",
    "        # return dist\n",
    "        return gaussian_kde(data)  # return the distribution\n",
    "\n",
    "    def laplace_smoothing(self, classes, length):\n",
    "        smooths = np.empty(self.total_classes, dtype=object)\n",
    "        for i in range(length):\n",
    "            smooths[i] = (classes[i] + self.laplace_smoothing_value) / \\\n",
    "                ((classes[i].size) +\n",
    "                 self.laplace_smoothing_value * self.n_features)\n",
    "\n",
    "        return smooths\n",
    "\n",
    "    def probability(self, data):\n",
    "        py = np.ones(self.total_classes)\n",
    "\n",
    "        for i in range(self.n_features):\n",
    "            for j in range(self.total_classes):\n",
    "                py[j] *= self.features['X'+str(i)+'-'+str(j)].pdf(data[i])\n",
    "\n",
    "        for i in range(self.total_classes):\n",
    "            py[i] *= self.prior[i]\n",
    "\n",
    "        # print('while training: ', py)\n",
    "        return py\n",
    "\n",
    "    def laplace_probability(self, data):\n",
    "\n",
    "        py = np.ones(self.total_classes)\n",
    "        pdf = np.ones(self.total_classes)\n",
    "\n",
    "        for i in range(self.n_features):\n",
    "            for j in range(self.total_classes):\n",
    "                pdf[j] *= self.features['X'+str(i)+'-'+str(j)].pdf(data[i])\n",
    "\n",
    "            smooths = self.laplace_smoothing(pdf, self.total_classes)\n",
    "            for j in range(self.total_classes):\n",
    "                py[j] *= smooths[j]\n",
    "\n",
    "        for i in range(self.total_classes):\n",
    "            py[i] *= self.prior[i]\n",
    "\n",
    "        return py\n",
    "\n",
    "    def fit(self):\n",
    "\n",
    "        X_training_class = np.empty(self.total_classes, dtype=object)\n",
    "        for i in range(self.total_classes):\n",
    "            X_training_class[i] = self.X_train[self.y_train == i]\n",
    "\n",
    "        self.prior = np.zeros(self.total_classes)\n",
    "        for i in range(self.total_classes):\n",
    "            # print('setting prior', i)\n",
    "            # if self.for_arc or self.for_plm:\n",
    "            #     self.prior[i] = 1 - (len(X_training_class[i]) / len(self.X_train))\n",
    "            # else:\n",
    "            #     self.prior[i] = len(X_training_class[i]) / len(self.X_train)\n",
    "            self.prior[i] = len(X_training_class[i]) / len(self.X_train)\n",
    "\n",
    "        self.n_features = self.X_train.shape[1]\n",
    "        print('Number of features: ', self.n_features)\n",
    "        self.features = {}\n",
    "\n",
    "        for i in range(self.n_features):\n",
    "            for j in range(self.total_classes):\n",
    "                temp = 'X'+str(i)+'-'+str(j)\n",
    "                self.features[temp] = self.fit_distribution(X_training_class[j][:, i])\n",
    "                # print(str(i), str(j), temp)\n",
    "                # self.list_of_ranges.append(temp)\n",
    "                # if temp in self.set_of_ranges:\n",
    "                #     print(str(i), str(j), 'Error: ', temp, ' already exists')\n",
    "                # self.set_of_ranges.add(temp)\n",
    "\n",
    "\n",
    "    def newvaluate(self):\n",
    "        print(self.list_of_ranges)\n",
    "        print(self.set_of_ranges)\n",
    "\n",
    "    def predict(self, X_test, y_test):\n",
    "        y_predicted = []\n",
    "        \n",
    "        y = []\n",
    "        for sample, target in zip(X_test, y_test):\n",
    "\n",
    "            if self.laplace:\n",
    "                py = self.laplace_probability(sample)\n",
    "\n",
    "            else:\n",
    "                py = self.probability(sample)\n",
    "\n",
    "            # print(\"Model predicted class {} and truth was {}\".format(np.argmax(py), target))\n",
    "            # print(py)\n",
    "            # print(np.argmax(py))\n",
    "            y_predicted.append(np.argmax(py))\n",
    "            # print(py)\n",
    "            y.append(target)\n",
    "\n",
    "        y_predicted = np.array(y_predicted)\n",
    "        y = np.array(y)\n",
    "        return y_predicted, y\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "def draw_matrix(true_positive, false_positive, false_negative, true_negative):\n",
    "    '''\n",
    "      Draw a confusion matrix.\n",
    "    '''\n",
    "    matrix = np.array([[true_positive, false_negative],\n",
    "                      [false_positive, true_negative]])\n",
    "    plt.imshow(matrix, interpolation='nearest', cmap=plt.cm.viridis_r)\n",
    "    plt.colorbar()\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "def confusion_matrix_calculator(y, y_predicted, desired_class):\n",
    "    '''\n",
    "      Takes original classes and predicted classes as input\n",
    "    '''\n",
    "    print('total y', y.size)\n",
    "    print('total y_predicted', y_predicted.size)\n",
    "\n",
    "    actual = (y == desired_class)\n",
    "    predicted = (y_predicted == desired_class)\n",
    "    print('actual', actual.sum())\n",
    "    true_positive = (actual & predicted).sum()\n",
    "    false_negative = (actual & ~predicted).sum()\n",
    "    false_positive = (~actual & predicted).sum()\n",
    "    true_negative = (~actual & ~predicted).sum()\n",
    "\n",
    "    return true_positive, false_negative, false_positive, true_negative\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def validation(given_y, y_predicted, classification_class='Kingdom', desired_classes = classes_dict):\n",
    "    '''\n",
    "      Takes sampling technique as input\n",
    "\n",
    "      Returns:\n",
    "        precision, recall, accuracy\n",
    "    '''\n",
    "    for class_name in desired_classes.keys():\n",
    "        print(class_name)\n",
    "        print(dataset[dataset[classification_class] == int(\n",
    "            desired_classes.get(class_name))][classification_class].count())\n",
    "        true_positive, false_negative, false_positive, true_negative = confusion_matrix_calculator(y=given_y, y_predicted=y_predicted, desired_class=int(desired_classes.get(class_name)))\n",
    "        draw_matrix(true_positive= true_positive, true_negative= true_negative, false_positive= false_positive, false_negative= false_negative)\n",
    "        print(\"true positive:\", true_positive)\n",
    "        print(\"false negative:\", false_negative)\n",
    "        print(\"false positive:\", false_positive)\n",
    "        print(\"true negative:\", true_negative)\n",
    "        print(\"Precision:\", true_positive / (true_positive + false_positive))\n",
    "        print(\"True Positive rate or Recall:\",\n",
    "              true_positive / (true_positive + false_negative))\n",
    "        print(\"specificity, selectivity or True Negative Rate:\",\n",
    "              true_negative / (true_negative + false_positive))\n",
    "        print(\"Accuracy:\", (true_positive + true_negative) /\n",
    "              (true_positive + false_positive + false_negative + true_negative))\n",
    "        print(\"\\n\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "def original_accuracy(y, y_predicted):\n",
    "    '''\n",
    "      Takes original classes and predicted classes as input\n",
    "    '''\n",
    "    actual = (y == y_predicted)\n",
    "    return actual.sum() / len(y)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using Sampling Technique:  smote\n",
      "X_training shape:  (25687, 64)\n",
      "y_training shape:  (25687,)\n",
      "X_testing shape:  (2606, 64)\n",
      "y_testing shape:  (2606,)\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Number of features:  64\n",
      "Model is trained\n",
      "X_testing shape:  (2606, 64)\n",
      "y_testing shape:  (2606,)\n",
      "Y_predicted:  [10 10 10 ... 10  6  9]\n",
      "y_predicted shape:  (2606,)\n",
      "Y_test:  [ 5  5 10 ...  1  6  4]\n",
      "y_test shape:  (2606,)\n",
      "Original Accuracy:  0.6289332310053722\n",
      "arc\n",
      "126\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 27\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 18\n",
      "false negative: 9\n",
      "false positive: 21\n",
      "true negative: 2558\n",
      "Precision: 0.46153846153846156\n",
      "True Positive rate or Recall: 0.6666666666666666\n",
      "specificity, selectivity or True Negative Rate: 0.9918573090345095\n",
      "Accuracy: 0.9884881043745203\n",
      "\n",
      "\n",
      "bct\n",
      "2919\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 618\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 303\n",
      "false negative: 315\n",
      "false positive: 135\n",
      "true negative: 1853\n",
      "Precision: 0.6917808219178082\n",
      "True Positive rate or Recall: 0.49029126213592233\n",
      "specificity, selectivity or True Negative Rate: 0.932092555331992\n",
      "Accuracy: 0.827321565617805\n",
      "\n",
      "\n",
      "phg\n",
      "220\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 37\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 29\n",
      "false negative: 8\n",
      "false positive: 108\n",
      "true negative: 2461\n",
      "Precision: 0.2116788321167883\n",
      "True Positive rate or Recall: 0.7837837837837838\n",
      "specificity, selectivity or True Negative Rate: 0.9579602958349552\n",
      "Accuracy: 0.955487336914812\n",
      "\n",
      "\n",
      "plm\n",
      "18\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 2\n",
      "false negative: 0\n",
      "false positive: 28\n",
      "true negative: 2576\n",
      "Precision: 0.06666666666666667\n",
      "True Positive rate or Recall: 1.0\n",
      "specificity, selectivity or True Negative Rate: 0.989247311827957\n",
      "Accuracy: 0.9892555640828856\n",
      "\n",
      "\n",
      "pln\n",
      "2523\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 464\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 213\n",
      "false negative: 251\n",
      "false positive: 108\n",
      "true negative: 2034\n",
      "Precision: 0.6635514018691588\n",
      "True Positive rate or Recall: 0.45905172413793105\n",
      "specificity, selectivity or True Negative Rate: 0.9495798319327731\n",
      "Accuracy: 0.8622409823484267\n",
      "\n",
      "\n",
      "inv\n",
      "1345\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 298\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 154\n",
      "false negative: 144\n",
      "false positive: 145\n",
      "true negative: 2163\n",
      "Precision: 0.5150501672240803\n",
      "True Positive rate or Recall: 0.5167785234899329\n",
      "specificity, selectivity or True Negative Rate: 0.9371750433275563\n",
      "Accuracy: 0.8891020721412126\n",
      "\n",
      "\n",
      "vrt\n",
      "2077\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 390\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 274\n",
      "false negative: 116\n",
      "false positive: 14\n",
      "true negative: 2202\n",
      "Precision: 0.9513888888888888\n",
      "True Positive rate or Recall: 0.7025641025641025\n",
      "specificity, selectivity or True Negative Rate: 0.9936823104693141\n",
      "Accuracy: 0.9501151189562548\n",
      "\n",
      "\n",
      "mam\n",
      "572\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 96\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 73\n",
      "false negative: 23\n",
      "false positive: 37\n",
      "true negative: 2473\n",
      "Precision: 0.6636363636363637\n",
      "True Positive rate or Recall: 0.7604166666666666\n",
      "specificity, selectivity or True Negative Rate: 0.9852589641434263\n",
      "Accuracy: 0.9769762087490407\n",
      "\n",
      "\n",
      "rod\n",
      "215\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 37\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 24\n",
      "false negative: 13\n",
      "false positive: 5\n",
      "true negative: 2564\n",
      "Precision: 0.8275862068965517\n",
      "True Positive rate or Recall: 0.6486486486486487\n",
      "specificity, selectivity or True Negative Rate: 0.9980537173997664\n",
      "Accuracy: 0.9930928626247122\n",
      "\n",
      "\n",
      "pri\n",
      "180\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 38\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUkAAAD8CAYAAAD6+lbaAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAAAaLUlEQVR4nO3df5Bd5X3f8fdHiySP4zQINhGyUEGuldQwDgJ2JDJkYmx+CaaDyMR1RFsjXBi5GWh+dwphBjyizMhpYlKPMfYGVITjACqxzaaVQ4WAoRkC1trFCESx1sIUbQQyiGJnoFBJ3/5xnmsfVvfcPbv3nL3nXj4vz5k95zk/7nOsmS/Pc55figjMzKy9eb3OgJlZkzlImpl14CBpZtaBg6SZWQcOkmZmHThImpl10FWQlHScpO2S9qS/iwquOyzpybSN5dKXS3pC0oSkeyUt6CY/ZmZV67YkeS2wIyJWADvScTtvRsTKtF2SS/8scEtEfBB4Dbiyy/yYmVWq2yC5FtiS9rcAl5a9UZKAjwH3zeZ+M3v3kLRM0sOSdkt6RtLvpPTPSJrM1VQvzt1zXaqlPifpwlz6mpQ2IamoYPcTx3SZ98URsT/tvwQsLrjuPZLGgUPApoj4BnA88H8i4lC6Zh+wtOiHJG0ANgD8zHt15i99cH6XWbe5JNTrLNgM/ODF/8crBw939Y82rBPibd4ude2Pee2BiFjT4ZJDwB9ExHck/SzwbUnb07lbIuJP8hdLOgVYB5wKvB94UNIvptO3AueTxZydksYiYnfRD08bJCU9CJzQ5tT1+YOICElFYxxPiohJSR8AHpK0C3h9ut+e8vxRYBTgzNMWxmN/UxhPrYHma6jXWbAZWHXhi10/423eZvW880td++CRrcOdzqfC2P60/2NJz9KhUEVWy70nIt4Cnpc0AaxK5yYiYi+ApHvStbMPkhFxXtE5SS9LWhIR+yUtAQ4UPGMy/d0r6RHgdOCvgGMlHZNKkycCk9Plx8z6h+aVLIweYTjVNltGU8Ho6GdKJ5PFkCeAs4FrJF0OjJOVNl8jC6CP527L11RfnJK+ulPWuv0mOQasT/vrgfunXiBpkaSFaX+Y7KV2RzazxsPAxzvdb2Z9SgLNK7fBKxExktuKAuT7yApYvxsRPwJuA/4JsJKspPmnVb9Gt0FyE3C+pD3AeekYSSOSbk/XfAgYl/RdsqC4KVf///fA76ei8PHAHV3mx8waQmQlyTJbqedJ88kC5Fcj4msAEfFyRByOiCPAn/PTKvUksCx3e6umWpReqKuGm4h4FTi3Tfo4cFXafwz4cMH9e/npS5nZIBEwVM236NQb5g7g2Yj4XC59Sa7x+NeBp9P+GPCXkj5H1nCzAvhWytUKScvJguM64F90+u1uW7fNzAooq3JX42zgk8AuSU+mtD8CLpO0EgjgB8CnASLiGUlbyRpkDgFXR8RhAEnXAA8AQ8DmiHim0w87SJpZbTSvmpHPEfG30LYf2bYO99wM3NwmfVun+6ZykDSz+lRXkuwZB0kzq4cEFZUke8lB0szq45KkmVkBgSpq3e4lB0kzq4mg7IibBnOQNLP6uLptZtaBg6SZWQHh1m0zs2LuAmRm1pmr22ZmBYRbt83MOnJJ0sysiFoT6vY1B0kzq4er22Zm03DrtplZkUon3e2ZrsK8pOMkbZe0J/1d1OaalZL+Li0o/pSk38ydu1PS87mFxVd2kx8za5BWdbvM1mDdloWvBXZExApgRzqe6g3g8og4FVgD/JmkY3Pn/11ErEzbk13mx8yaRCq3NVi3QXItsCXtbwEunXpBRHwvIvak/b8nW5v757v8XTPrAyGV2pqs2yC5OLdS2UvA4k4XS1oFLAC+n0u+OVXDb2mtz21mA2Jeya3Bpm24kfQgcEKbU9fnDyIiJEWH5ywBvgKsT2vkAlxHFlwXAKNk63BvLLh/A7ABYNnS/p/I02zgvVuWb4iI84rOSXq5te5tCoIHCq77R8B/A66PiMdzz26VQt+S9J+BP+yQj1GyQMqZpy0sDMZm1iANr0qX0W2YHwPWp/31wP1TL5C0APg6cFdE3Dfl3JL0V2TfM5+eer+Z9S9/k4RNwPmS9gDnpWMkjUi6PV3zCeDXgCvadPX5qqRdwC5gGPgPXebHzJpCvDu+SXYSEa8C57ZJHweuSvt/AfxFwf0f6+b3zazhGl5KLMMjbsysFoGId0PDjZnZrLSq233OQdLM6uPqtplZsaa3XJfhIGlm9en/GOkgaWY1cknSzKyAIIYcJM3MCkX/x0gHSTOrS/PniixjAHoxmVkjaQbbdI+Slkl6WNLutMrB76T0tqsjKPN5SRNpKsYzcs9an67fI2l90W+2OEiaWW0qnODiEPAHEXEKcBZwtaRTKF4d4SJgRdo2ALdBFlSBG4HVwCrgxnbLzuQ5SJpZfSqa4CIi9kfEd9L+j4FngaUUr46wlmzmsUjTMx6bZh27ENgeEQcj4jVgO9myMoX8TdLMahGCqGGRL0knA6cDT1C8OsJS4MXcbftSWlF6IQdJM6vNDFq3hyWN545H00Tb7yDpfcBfAb8bET9Srqo+3eoIs+UgaWb1Kd+6/UpEjHR+lOaTBcivRsTXUnLR6giTwLLc7SemtEngnCnpj3T6XX+TNLPahMpt00mrF9wBPBsRn8udKlodYQy4PLVynwW8nqrlDwAXSFqUGmwuSGmFXJI0s/pU90nybOCTwC5JT6a0PyJbDWGrpCuBF8hWQgDYBlwMTABvAJ8CiIiDkm4CdqbrNkbEwU4/7CBpZvWocFhiRPwtxSG33eoIAVxd8KzNwOayv+0gaWb16f8BN9V8k5S0RtJzqXf7tW3OL5R0bzr/RGrCb527LqU/J+nCKvJjZs1Q1TfJXuo6SEoaAm4l6+F+CnBZ6gmfdyXwWkR8ELgF+Gy69xRgHXAqWYfOL6bnmVm/k8pvDVZFSXIVMBEReyPibeAest7uefle8fcB56bWqrXAPRHxVkQ8T/aRdVUFeTKzHgtckmwp04P9J9dExCHgdeD4kvcCIGmDpHFJ4z989XAF2Taz2lU0wUUv9U0/yYgYjYiRiBj5+eNdIzdrvDQssczWZFUEyaKe7W2vkXQM8HPAqyXvNbM+5ep2ZiewQtJySQvIGmLGplyT7xX/ceCh1I9pDFiXWr+Xk01r9K0K8mRmTTAA1e2u+0lGxCFJ15AN7RkCNkfEM5I2AuMRMUY2nOgrkiaAg2SBlHTdVmA32XxxV0eEPziaDYjomw96xSrpTB4R28iGAeXTbsjt/1/gnxfcezNwcxX5MLMG6YNSYhkecWNmtWn698YyHCTNrBZBPZPuzjUHSTOrT//HSAdJM6tJH3TvKcNB0szq4yBpZlbMJUkzsw7cT9LMrIj7SZqZdRYNnyuyDAdJM6tP/8dIB0kzq0dr0t1+5yBpZvVxkDQzKyC3bpuZdeaSpJlZMX+TNDPrxEHSzKyAJ7gwM5vGAATJStqeJK2R9JykCUnXtjn/+5J2S3pK0g5JJ+XOHZb0ZNqmLiBmZn0s5pXbmqzrkqSkIeBW4HxgH7BT0lhE7M5d9j+BkYh4Q9JvAX8M/GY692ZErOw2H2bWQC5JArAKmIiIvRHxNnAPsDZ/QUQ8HBFvpMPHydbXNrMBFjPYmqyKILkUeDF3vC+lFbkS+Gbu+D2SxiU9LunSopskbUjXjf/wVa86a9Z4Zdfcbnhpc04bbiT9K2AE+Egu+aSImJT0AeAhSbsi4vtT742IUWAU4MzTFjb9Pz5mBo0PgGVUESQngWW54xNT2jtIOg+4HvhIRLzVSo+IyfR3r6RHgNOBo4KkmfWfpjfKlFHFK+wEVkhaLmkBsA54Ryu1pNOBLwOXRMSBXPoiSQvT/jBwNpBv8DGzPhYqtzVZ10EyIg4B1wAPAM8CWyPiGUkbJV2SLvuPwPuA/zKlq8+HgHFJ3wUeBjZNaRU3s35V4TdJSZslHZD0dC7tM5Imc10IL86duy51SXxO0oW59I7dFdup5JtkRGwDtk1JuyG3f17BfY8BH64iD2bWQNWVEu8EvgDcNSX9loj4k3f8pHQKWY32VOD9wIOSfjGdnq674lE84sbMalNVVToiHpV0csnL1wL3pLaP5yVNkHVVhNRdEUBSq7tixyA5AJ9Vzayxyle3h1td/NK2oeQvXJNG8m2WtCilFXVLnGl3RcAlSTOry8wm3X0lIkZm+Au3ATeR9Ue/CfhT4F/P8BnTcpA0s/rU2HIdES//5GekPwf+azrs1C1x2u6KU7m6bWa1aC0EVlcXIElLcoe/DrRavseAdZIWSloOrAC+RYnuiu24JGlm9amoJCnpbuAcsm+X+4AbgXMkrSSLxz8APg2QuiBuJWuQOQRcHRGH03Na3RWHgM0R8cx0v+0gaWb1qa51+7I2yXd0uP5m4OY26Ud1V5yOg6SZ1WNAVkscgFcwM6uPS5JmVpumj8suw0HSzOrjIGlm1oGDpJlZgT6YBq0MB0kzq80gtG47SJpZfVySNDPrQP2/HJWDpJnVoh+WZijDQdLM6jMAQbKSz6rTrRsh6QpJP8ytRXFV7tx6SXvStr6K/JhZQ3jdbZA0RLl1I+6NiGum3Hsc2WweI2QzeXw73ftat/kys94bhNbtKl5hFWndiIh4G2itG1HGhcD2iDiYAuN2YE0FeTKzXqtwtcRequKbZLt1I1a3ue43JP0a8D3g9yLixYJ72645kda82ACwbOkQRzhSQdZtrlz4/jN7nQWbge/Fq9U8aABat+eqMPzXwMkR8ctkpcUtM31ARIxGxEhEjAwfP1R5Bs2senXOTD5XqgiSndaTACAiXk3LOwLcDpxZ9l4zs16qIkhOu27ElLUoLgGeTfsPABdIWpSWg7wgpZnZAIh55bYm6/qbZEQcarduhKSNwHhEjAG/LekSsvUmDgJXpHsPSrqJLNACbIyIg93mycwaQAzEN8lKOpO3WzciIm7I7V8HXFdw72ZgcxX5MLOGafj3xjI84sbM6uMgaWbWgYOkmVmR8DdJM7NCA7KkrIOkmdXHJUkzsw78TdLMrJgcJM3MCrgzuZnZNFySNDPrwCVJM7MOXJI0MysgkEuSZmYduCRpZlYkBqIkOQCDhsysqTSv3Dbtc6TNkg5IejqXdpyk7Wk56u1p4m6U+Xxa4vopSWfk7pnxEtYOkmZWj1Y/yTLb9O7k6JVUrwV2RMQKYEc6BrgIWJG2DcBt8I4lrFeTrfJ6YyuwduIgaWa1EFnDTZltOhHxKNmqBnlr+emigluAS3Ppd0XmceDYtITMrJaw9jdJM6vNDIYlDksazx2PRsToNPcsjoj9af8lYHHaL1qquvQS1nmVBElJa4D/RLbGze0RsWnK+VuAj6bD9wK/EBHHpnOHgV3p3P+OiEuqyJOZ9d4MGm5eiYiR2f5ORIRqaiXqOkhKGgJuBc4ni8w7JY1FxO7WNRHxe7nr/y1weu4Rb0bEym7zYWYNVG8XoJclLYmI/ak6fSClFy1VPQmcMyX9kel+pIpvkquAiYjYGxFvA/eQfRMochlwdwW/a2ZNJpinKLXN0hjQaqFeD9yfS788tXKfBbyequWzWsK6iup2u3r+6nYXSjoJWA48lEt+T/oWcQjYFBHfKLh3A1lLFcuWDnWfazOrlSrsJynpbrJS4LCkfWSt1JuArZKuBF4APpEu3wZcDEwAbwCfgtkvYT3XDTfrgPsi4nAu7aSImJT0AeAhSbsi4vtTb0wfcUcBzjhtYf/3UDV7F6gqSEbEZQWnzm1zbQBXFzxnxktYV1HdLqr/t7OOKVXtiJhMf/eSfR84/ejbzKwfSeW2JqsiSO4EVkhaLmkBWSAcm3qRpH8KLAL+Lpe2SNLCtD8MnA3snnqvmfWnmr9Jzomuq9sRcUjSNWQfQIeAzRHxjKSNwHhEtALmOuCeVBRu+RDwZUlHyAL2pnyruJn1L3kWoJ+KiG1kH0vzaTdMOf5Mm/seAz5cRR7MrHnmzXOQNDMr0PyqdBkOkmZWi9bY7X7nIGlmtZmHg6SZWXt90L2nDAdJM6uFgKF5R3qdja45SJpZTdxwY2bWkRtuzMwKKM0C1O8cJM2sNm7dNjPrwNVtM7MCIjjGrdtmZu0Jf5M0Myvmhhszs87ccGNmVkDuTG5mVkzghhszs05ckjQzKyAG45tkFQuBIWmzpAOSni44L0mflzQh6SlJZ+TOrZe0J23r291vZn2o5CJgTS9tVhIkgTuBNR3OXwSsSNsG4DYASceRLTK+GlgF3ChpUUV5MrMem6cjpbYmqyRIRsSjwMEOl6wF7orM48CxkpYAFwLbI+JgRLwGbKdzsDWzPtHqTN7vJcm5+ia5FHgxd7wvpRWlH0XSBrJSKMuWDtWTSzOrjAiOaXgpsYyqqtu1i4jRiBiJiJHh4x0kzfqBq9vlTQLLcscnprSidDPrcxIMKUptTTZXQXIMuDy1cp8FvB4R+4EHgAskLUoNNhekNDMbAPOIUluTVfJNUtLdwDnAsKR9ZC3W8wEi4kvANuBiYAJ4A/hUOndQ0k3AzvSojRHRqQHIzPpGNL4qXUYlQTIiLpvmfABXF5zbDGyuIh9m1hyeKs3MrAMRzNfhXmejaw6SZlYblyTNzAq0Wrf7Xd/0kzSz/jOPI6W2MiT9QNIuSU9KGk9px0nanuZ+2N4a1txpvoiZv4OZWQ1ak+5WPCzxoxGxMiJG0vG1wI6IWAHsSMdQMF/EbDhImlktBMzX4VJbF9YCW9L+FuDSXHq7+SJmzEHSzGoSDOlIqY2sj/V4btvQ9oHw3yV9O3d+cRqYAvASsDjtl54XYjpuuDGzWsxw0t1XclXoIr8aEZOSfgHYLul/5U9GREjVtxQ5SJpZbYYqHHETEZPp7wFJXyebg/ZlSUsiYn+qTh9Il1c2L4Sr22ZWCykqmwVI0s9I+tnWPtk8D0+TzQvRWtFgPXB/2i+aL2LGXJI0s9oMVTd5xWLg65Igi1t/GRF/I2knsFXSlcALwCfS9W3ni5gNB0kzq0WVwxIjYi9wWpv0V4Fz26QXzhcxUw6SZlaLbIILzwJkZlZoqORomiZzkDSzWrRG3PQ7B0kzq41LkmZmBUS1/SR7xUHSzGohDcaku5V0Jpe0WdIBSU8XnP+XabqiXZIek3Ra7txR0x+Z2WCocqq0XqmqJHkn8AXgroLzzwMfiYjXJF0EjAKrc+c/GhGvVJQXM2sA0fzlYsuoaiGwRyWd3OH8Y7nDx8nGUZrZABOD0XDTi7HbVwLfzB23m/7IzAZAVWO3e2lOG24kfZQsSP5qLvmo6Y8i4tE2924gm2GYZUuH5iS/ZjZ7IljghpvyJP0ycDuwNo23BN45/RHQmv7oKBExGhEjETEyfLyDpFnTZfNJ9n/DzZwESUn/GPga8MmI+F4uvWj6IzPrezOambyxKqluS7obOIdsCvZ9wI3AfICI+BJwA3A88MU01dGhNAtx2+mPqsiTmfXWoDTcVNW6fdk0568CrmqT3nb6IzMbDDNYvqGxPOLGzGohReOr0mU4SJpZLQTMp/9btx0kzawmLkmamRXKGm78TdLMrJAn3TUzKyDCXYDMzIq4um1m1oGA+W64MTMr5pKkmVmBbN3tXueiew6SZlaTcEnSzKxINlVa/3OQNLNaSGKB+r++7SBpZrVxSdLMrICAIZckzcyKzcNB0sysrWzEjYOkmVkhlyTNzAoIMV/9v7JpJY1PkjZLOiCp7UqHks6R9LqkJ9N2Q+7cGknPSZqQdG0V+TGz3sv6SarU1mRVlSTvBL4A3NXhmv8REf8snyBpCLgVOB/YB+yUNBYRuyvKl5n1jBhS/3cCquQNIuJR4OAsbl0FTETE3oh4G7gHWFtFnsyst7KSZLn/NdlcfpP8FUnfBf4e+MOIeAZYCryYu2YfsLrdzZI2ABvS4Vvvff8Lbav2fW4YeKXXmajHC4P6boP6Xr/U7QO+/dRbDwwt2TNc8vLG/n84V0HyO8BJEfEPki4GvgGsmMkDImIUGAWQNB4RI5XnsscG9b1gcN9tkN+r22dExJoq8tJrc1LOjYgfRcQ/pP1twHxJw8AksCx36YkpzcysEeYkSEo6QcrGJ0lalX73VWAnsELSckkLgHXA2FzkycysjEqq25LuBs4BhiXtA24E5gNExJeAjwO/JekQ8CawLiICOCTpGuABYAjYnL5VTme0inw30KC+Fwzuu/m9BpyyWGVmZu00u+3dzKzHHCTNzDroiyAp6ThJ2yXtSX8XFVx3ODf0sbENQNMNxZS0UNK96fwTkk7uQTZnrMR7XSHph7l/o6t6kc+ZKjHsVpI+n977KUlnzHUeZ6Ob4cTvJn0RJIFrgR0RsQLYkY7beTMiVqbtkrnLXnm5oZgXAacAl0k6ZcplVwKvRcQHgVuAz85tLmeu5HsB3Jv7N7p9TjM5e3cCnfr8XUTW73cF2YCH2+YgT1W4k87vBdlw4ta/18Y5yFPj9EuQXAtsSftbgEt7l5WulRmKmX/f+4BzW12oGmxgh5iWGHa7FrgrMo8Dx0paMje5m70uhhO/q/RLkFwcEfvT/kvA4oLr3iNpXNLjki6dm6zNWLuhmEuLromIQ8DrwPFzkrvZK/NeAL+RqqT3SVrW5nw/Kvvu/ehXJH1X0jclndrrzPRCY+aTlPQgcEKbU9fnDyIiJBX1WzopIiYlfQB4SNKuiPh+1Xm1Wftr4O6IeEvSp8lKyx/rcZ6sWNfDiQdBY4JkRJxXdE7Sy5KWRMT+VI05UPCMyfR3r6RHgNOBpgXJMkMxW9fsk3QM8HNkI5SabNr3ioj8O9wO/PEc5GsuDOTw2oj4UW5/m6QvShqOiMZORlGHfqlujwHr0/564P6pF0haJGlh2h8GzgaaOC9lmaGY+ff9OPBQNL/X/7TvNeU73SXAs3OYvzqNAZenVu6zgNdzn4f6VofhxO8qjSlJTmMTsFXSlcALwCcAJI0A/yYirgI+BHxZ0hGyf8xNTZy8NyLaDsWUtBEYj4gx4A7gK5ImyD6sr+tdjssp+V6/LekS4BDZe13RswzPQIlht9uAi4EJ4A3gU73J6cx0MZz4XcXDEs3MOuiX6raZWU84SJqZdeAgaWbWgYOkmVkHDpJmZh04SJqZdeAgaWbWwf8H7Kj8gYRCMuAAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 29\n",
      "false negative: 9\n",
      "false positive: 46\n",
      "true negative: 2522\n",
      "Precision: 0.38666666666666666\n",
      "True Positive rate or Recall: 0.7631578947368421\n",
      "specificity, selectivity or True Negative Rate: 0.9820872274143302\n",
      "Accuracy: 0.9788948580199539\n",
      "\n",
      "\n",
      "vrl\n",
      "2831\n",
      "total y 2606\n",
      "total y_predicted 2606\n",
      "actual 599\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "true positive: 520\n",
      "false negative: 79\n",
      "false positive: 320\n",
      "true negative: 1687\n",
      "Precision: 0.6190476190476191\n",
      "True Positive rate or Recall: 0.8681135225375626\n",
      "specificity, selectivity or True Negative Rate: 0.8405580468360737\n",
      "Accuracy: 0.8468917881811205\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# techniques = ['smote', 'ros']\n",
    "techniques = ['smote']\n",
    "\n",
    "for technique in techniques:\n",
    "    # sampler(X, y, technique)\n",
    "    print('Using Sampling Technique: ', technique)\n",
    "    SCALLED_X_train, SCALLED_y_train = sampler(X, y ,technique=technique)\n",
    "    data_generator = DataGenerator(y=SCALLED_y_train, X=SCALLED_X_train)\n",
    "    X_train, X_test, y_train, y_test = data_generator.split_data()\n",
    "    # X_train, y_train, \n",
    "    # X_train, y_train = X_training, y_training\n",
    "    print('X_training shape: ', X_train.shape)\n",
    "    print('y_training shape: ', y_train.shape)\n",
    "    print('X_testing shape: ', X_testing.shape)\n",
    "    print('y_testing shape: ', y_testing.shape)\n",
    "\n",
    "    print('\\n')\n",
    "    y_train.value_counts().plot(kind='bar')\n",
    "    plt.show()\n",
    "    print('\\n')\n",
    "\n",
    "    clf = MultiClassClassification(X_train=X_train, y_train=y_train, total_classes=11, laplace=False, laplace_smoothing_value=0.5,  for_arc=False, for_plm=False)\n",
    "    clf.fit()\n",
    "    print('Model is trained')\n",
    "    # y_predicted, y\n",
    "    y_predicted, given_y = clf.predict(X_test=X_testing, y_test=y_testing)\n",
    "    print('X_testing shape: ', X_testing.shape)\n",
    "    print('y_testing shape: ', y_testing.shape)\n",
    "    \n",
    "    print('Y_predicted: ', y_predicted)\n",
    "    print('y_predicted shape: ', y_predicted.shape)\n",
    "    print('Y_test: ', given_y)\n",
    "    print('y_test shape: ', given_y.shape)\n",
    "    print('Original Accuracy: ', original_accuracy(given_y, y_predicted))\n",
    "    validation(given_y= given_y, y_predicted= y_predicted, desired_classes = classes_dict)\n",
    "\n",
    "\n",
    "    # 'arc': 0, 'bct': 1, 'phg': 2, 'plm': 3, \n",
    "    # clf_arc = MultiClassClassification(X_train=X_training, y_train=y_training, total_classes=11, laplace=False, laplace_smoothing_value=0.5, for_arc=True)\n",
    "    # clf_arc.fit()\n",
    "    # print('Model is trained for ARC')\n",
    "    # y_predicted, given_y = clf_arc.predict(X_test=X_testing, y_test=y_testing)\n",
    "    # print('Original Accuracy: ', original_accuracy(given_y, y_predicted))\n",
    "    # validation(given_y= given_y, y_predicted= y_predicted, desired_classes = classes_dict)\n",
    "    \n",
    "    \n",
    "    # print('for prc')\n",
    "    # clf_prc = MultiClassClassification(X_train=X_training, y_train=y_training, laplace=False, total_classes=11, laplace_smoothing_value=0.5, for_plm=True)\n",
    "    # clf_prc.fit()\n",
    "    # y_predicted, given_y = clf_prc.predict(X_test=X_testing, y_test=y_testing)\n",
    "    # print('Original Accuracy: ', original_accuracy(given_y, y_predicted))\n",
    "    # validation(given_y= given_y, y_predicted= y_predicted, desired_classes = classes_dict)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.12 ('.env': venv)",
   "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": "3.9.12"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "20f3b51db45487e76e19b963685aeaff4d2f53ee3a54c0a9156da0ef575c3172"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}