project-Replikacija-Rezultata.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cde4f089-0c77-4965-9b1b-95247d2a4fae",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "90a87d05-dd7d-44b3-94f6-61ad077ca455",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder, LabelEncoder\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.feature_selection import SelectKBest, mutual_info_classif\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import precision_score,recall_score,f1_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import roc_curve, roc_auc_score\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.svm import SVC\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.ensemble import VotingClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.utils import resample\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.ensemble import GradientBoostingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b3fcab79-571b-4172-a995-33cb99ecc438",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c5b9e8bf-fff1-4c2d-9e55-cfa79915146e",
   "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>Subject ID</th>\n",
       "      <th>Group</th>\n",
       "      <th>Visit</th>\n",
       "      <th>MR Delay</th>\n",
       "      <th>M/F</th>\n",
       "      <th>Age</th>\n",
       "      <th>EDUC</th>\n",
       "      <th>SES</th>\n",
       "      <th>MMSE</th>\n",
       "      <th>CDR</th>\n",
       "      <th>eTIV</th>\n",
       "      <th>nWBV</th>\n",
       "      <th>ASF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>OAS2_0001</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>87</td>\n",
       "      <td>14</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0.696</td>\n",
       "      <td>0.883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>OAS2_0001</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>2</td>\n",
       "      <td>457</td>\n",
       "      <td>M</td>\n",
       "      <td>88</td>\n",
       "      <td>14</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2004</td>\n",
       "      <td>0.681</td>\n",
       "      <td>0.876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>OAS2_0002</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>75</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1678</td>\n",
       "      <td>0.736</td>\n",
       "      <td>1.046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>OAS2_0002</td>\n",
       "      <td>Demented</td>\n",
       "      <td>2</td>\n",
       "      <td>560</td>\n",
       "      <td>M</td>\n",
       "      <td>76</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1738</td>\n",
       "      <td>0.713</td>\n",
       "      <td>1.010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>OAS2_0002</td>\n",
       "      <td>Demented</td>\n",
       "      <td>3</td>\n",
       "      <td>1895</td>\n",
       "      <td>M</td>\n",
       "      <td>80</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1698</td>\n",
       "      <td>0.701</td>\n",
       "      <td>1.034</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366</th>\n",
       "      <td>OAS2_0185</td>\n",
       "      <td>Demented</td>\n",
       "      <td>2</td>\n",
       "      <td>842</td>\n",
       "      <td>M</td>\n",
       "      <td>82</td>\n",
       "      <td>16</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1693</td>\n",
       "      <td>0.694</td>\n",
       "      <td>1.037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>367</th>\n",
       "      <td>OAS2_0185</td>\n",
       "      <td>Demented</td>\n",
       "      <td>3</td>\n",
       "      <td>2297</td>\n",
       "      <td>M</td>\n",
       "      <td>86</td>\n",
       "      <td>16</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1688</td>\n",
       "      <td>0.675</td>\n",
       "      <td>1.040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>368</th>\n",
       "      <td>OAS2_0186</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>61</td>\n",
       "      <td>13</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1319</td>\n",
       "      <td>0.801</td>\n",
       "      <td>1.331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>OAS2_0186</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>2</td>\n",
       "      <td>763</td>\n",
       "      <td>F</td>\n",
       "      <td>63</td>\n",
       "      <td>13</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1327</td>\n",
       "      <td>0.796</td>\n",
       "      <td>1.323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>370</th>\n",
       "      <td>OAS2_0186</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>3</td>\n",
       "      <td>1608</td>\n",
       "      <td>F</td>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1333</td>\n",
       "      <td>0.801</td>\n",
       "      <td>1.317</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>371 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Subject ID        Group  Visit  MR Delay M/F  Age  EDUC       SES  MMSE  \\\n",
       "0    OAS2_0001  Nondemented      1         0   M   87    14  2.000000  27.0   \n",
       "1    OAS2_0001  Nondemented      2       457   M   88    14  2.000000  30.0   \n",
       "2    OAS2_0002     Demented      1         0   M   75    12  2.460452  23.0   \n",
       "3    OAS2_0002     Demented      2       560   M   76    12  2.460452  28.0   \n",
       "4    OAS2_0002     Demented      3      1895   M   80    12  2.460452  22.0   \n",
       "..         ...          ...    ...       ...  ..  ...   ...       ...   ...   \n",
       "366  OAS2_0185     Demented      2       842   M   82    16  1.000000  28.0   \n",
       "367  OAS2_0185     Demented      3      2297   M   86    16  1.000000  26.0   \n",
       "368  OAS2_0186  Nondemented      1         0   F   61    13  2.000000  30.0   \n",
       "369  OAS2_0186  Nondemented      2       763   F   63    13  2.000000  30.0   \n",
       "370  OAS2_0186  Nondemented      3      1608   F   65    13  2.000000  30.0   \n",
       "\n",
       "     CDR  eTIV   nWBV    ASF  \n",
       "0    0.0  1987  0.696  0.883  \n",
       "1    0.0  2004  0.681  0.876  \n",
       "2    0.5  1678  0.736  1.046  \n",
       "3    0.5  1738  0.713  1.010  \n",
       "4    0.5  1698  0.701  1.034  \n",
       "..   ...   ...    ...    ...  \n",
       "366  0.5  1693  0.694  1.037  \n",
       "367  0.5  1688  0.675  1.040  \n",
       "368  0.0  1319  0.801  1.331  \n",
       "369  0.0  1327  0.796  1.323  \n",
       "370  0.0  1333  0.801  1.317  \n",
       "\n",
       "[371 rows x 13 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"obradeni_podaci.csv\")\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "71047e55-8e28-41d3-83d2-d6743ab7b095",
   "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>Subject ID</th>\n",
       "      <th>Group</th>\n",
       "      <th>Visit</th>\n",
       "      <th>MR Delay</th>\n",
       "      <th>M/F</th>\n",
       "      <th>Age</th>\n",
       "      <th>EDUC</th>\n",
       "      <th>SES</th>\n",
       "      <th>MMSE</th>\n",
       "      <th>CDR</th>\n",
       "      <th>eTIV</th>\n",
       "      <th>nWBV</th>\n",
       "      <th>ASF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>OAS2_0001</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>87</td>\n",
       "      <td>14</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0.696</td>\n",
       "      <td>0.883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>OAS2_0002</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>75</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1678</td>\n",
       "      <td>0.736</td>\n",
       "      <td>1.046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>OAS2_0004</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>88</td>\n",
       "      <td>18</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1215</td>\n",
       "      <td>0.710</td>\n",
       "      <td>1.444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>OAS2_0005</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>80</td>\n",
       "      <td>12</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1689</td>\n",
       "      <td>0.712</td>\n",
       "      <td>1.039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>OAS2_0007</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>71</td>\n",
       "      <td>16</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1357</td>\n",
       "      <td>0.748</td>\n",
       "      <td>1.293</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>OAS2_0182</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>73</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1661</td>\n",
       "      <td>0.698</td>\n",
       "      <td>1.056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>OAS2_0183</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>66</td>\n",
       "      <td>13</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.746</td>\n",
       "      <td>1.174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>OAS2_0184</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>72</td>\n",
       "      <td>16</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>24.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1354</td>\n",
       "      <td>0.733</td>\n",
       "      <td>1.296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>OAS2_0185</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>80</td>\n",
       "      <td>16</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1704</td>\n",
       "      <td>0.711</td>\n",
       "      <td>1.030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>OAS2_0186</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>61</td>\n",
       "      <td>13</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1319</td>\n",
       "      <td>0.801</td>\n",
       "      <td>1.331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Subject ID        Group  Visit  MR Delay M/F  Age  EDUC       SES  MMSE  \\\n",
       "0    OAS2_0001  Nondemented      1         0   M   87    14  2.000000  27.0   \n",
       "1    OAS2_0002     Demented      1         0   M   75    12  2.460452  23.0   \n",
       "2    OAS2_0004  Nondemented      1         0   F   88    18  3.000000  28.0   \n",
       "3    OAS2_0005  Nondemented      1         0   M   80    12  4.000000  28.0   \n",
       "4    OAS2_0007     Demented      1         0   M   71    16  2.460452  28.0   \n",
       "..         ...          ...    ...       ...  ..  ...   ...       ...   ...   \n",
       "145  OAS2_0182     Demented      1         0   M   73    12  2.460452  23.0   \n",
       "146  OAS2_0183  Nondemented      1         0   F   66    13  2.000000  30.0   \n",
       "147  OAS2_0184     Demented      1         0   F   72    16  3.000000  24.0   \n",
       "148  OAS2_0185     Demented      1         0   M   80    16  1.000000  28.0   \n",
       "149  OAS2_0186  Nondemented      1         0   F   61    13  2.000000  30.0   \n",
       "\n",
       "     CDR  eTIV   nWBV    ASF  \n",
       "0    0.0  1987  0.696  0.883  \n",
       "1    0.5  1678  0.736  1.046  \n",
       "2    0.0  1215  0.710  1.444  \n",
       "3    0.0  1689  0.712  1.039  \n",
       "4    0.5  1357  0.748  1.293  \n",
       "..   ...   ...    ...    ...  \n",
       "145  0.5  1661  0.698  1.056  \n",
       "146  0.0  1495  0.746  1.174  \n",
       "147  0.5  1354  0.733  1.296  \n",
       "148  0.5  1704  0.711  1.030  \n",
       "149  0.0  1319  0.801  1.331  \n",
       "\n",
       "[150 rows x 13 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = data.groupby('Subject ID').first().reset_index()\n",
    "\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "55f82b46-5a8a-40d0-89b1-c438aa9db70a",
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Group'] = data['Group'].replace('Converted', 'Demented')\n",
    "#data = data[data['Group'] != 'Converted']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "82afa130-8c32-4d8b-b6cb-3f6ec2d96e5c",
   "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>Subject ID</th>\n",
       "      <th>Group</th>\n",
       "      <th>Visit</th>\n",
       "      <th>MR Delay</th>\n",
       "      <th>M/F</th>\n",
       "      <th>Age</th>\n",
       "      <th>EDUC</th>\n",
       "      <th>SES</th>\n",
       "      <th>MMSE</th>\n",
       "      <th>CDR</th>\n",
       "      <th>eTIV</th>\n",
       "      <th>nWBV</th>\n",
       "      <th>ASF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>OAS2_0001</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>87</td>\n",
       "      <td>14</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1987</td>\n",
       "      <td>0.696</td>\n",
       "      <td>0.883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>OAS2_0002</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>75</td>\n",
       "      <td>12</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1678</td>\n",
       "      <td>0.736</td>\n",
       "      <td>1.046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>OAS2_0004</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>F</td>\n",
       "      <td>88</td>\n",
       "      <td>18</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1215</td>\n",
       "      <td>0.710</td>\n",
       "      <td>1.444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>OAS2_0005</td>\n",
       "      <td>Nondemented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>80</td>\n",
       "      <td>12</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1689</td>\n",
       "      <td>0.712</td>\n",
       "      <td>1.039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>OAS2_0007</td>\n",
       "      <td>Demented</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>M</td>\n",
       "      <td>71</td>\n",
       "      <td>16</td>\n",
       "      <td>2.460452</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1357</td>\n",
       "      <td>0.748</td>\n",
       "      <td>1.293</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Subject ID        Group  Visit  MR Delay M/F  Age  EDUC       SES  MMSE  \\\n",
       "0  OAS2_0001  Nondemented      1         0   M   87    14  2.000000  27.0   \n",
       "1  OAS2_0002     Demented      1         0   M   75    12  2.460452  23.0   \n",
       "2  OAS2_0004  Nondemented      1         0   F   88    18  3.000000  28.0   \n",
       "3  OAS2_0005  Nondemented      1         0   M   80    12  4.000000  28.0   \n",
       "4  OAS2_0007     Demented      1         0   M   71    16  2.460452  28.0   \n",
       "\n",
       "   CDR  eTIV   nWBV    ASF  \n",
       "0  0.0  1987  0.696  0.883  \n",
       "1  0.5  1678  0.736  1.046  \n",
       "2  0.0  1215  0.710  1.444  \n",
       "3  0.0  1689  0.712  1.039  \n",
       "4  0.5  1357  0.748  1.293  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "db653ecb-c710-44b2-8160-5ef32b3218c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Subject ID    150\n",
       "Group           2\n",
       "Visit           1\n",
       "MR Delay        1\n",
       "M/F             2\n",
       "Age            34\n",
       "EDUC           12\n",
       "SES             6\n",
       "MMSE           13\n",
       "CDR             3\n",
       "eTIV          131\n",
       "nWBV           90\n",
       "ASF           129\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "69543add-ea72-4c49-8b38-141e20b1e209",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Nondemented' 'Demented']\n"
     ]
    }
   ],
   "source": [
    "unique_groups = data['Group'].unique()\n",
    "print(unique_groups[:3])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79c2900c-da45-4138-9a11-87b43fafa9d9",
   "metadata": {},
   "source": [
    "1.Stablo odlučivanja\n",
    "\r\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6260b60e-c651-4911-adb3-3a2dd6ac7cd7",
   "metadata": {},
   "source": [
    "U svom projektu koristim stablo odlučivanja za predviđanje demencije kod osoba na temelju njihovih osobnih karakteristika ili značajki. Stablo odlučivanja pomaže u analizi podataka i donošenju zaključaka o demenciji na temelju dostupnih informacija poput dobi, spola, obrazovanja ili drugih relevantnih značajki. Ovim pristupom možemo stvoriti model koji predviđa prisutnost demencije na temelju tih karakteristika."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff8464b8-5047-483f-aede-dfaa5367f9f2",
   "metadata": {},
   "source": [
    "Pretvorba kategoričkih vrijednosti u numeričke"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "19e1ef07-8f53-4e5c-bd3b-946264bbe9bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "label_encoder = LabelEncoder()\n",
    "data['M/F'] = label_encoder.fit_transform(data['M/F'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "32004a44-db85-480c-9f0e-9fa791fef895",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Podijela podataka na značajke (X) i oznake (y)\n",
    "selected_features = ['M/F', 'Age', 'EDUC', 'SES', 'MMSE', 'eTIV', 'nWBV', 'ASF', 'CDR']\n",
    "X = data[selected_features]  # Odabir značajki za X\n",
    "y = data['Group'] # Varijabla koju želim predviđati"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "624bb29e-68d5-49d7-b4c2-e8290966c040",
   "metadata": {},
   "source": [
    "Ovdje se koristi DecisionTreeClassifier iz Scikit-Learn-a za izgradnju modela stabla odlučivanja. Podaci se dijele na skup za učenje i testiranje pomoću train_test_split. Zatim se model trenira na skupu za učenje (fit) i koristi se za predviđanje (predict) na skupu za testiranje. Konačno, koristi se accuracy_score i confusion_matrix za evaluaciju performansi modela."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5395e9fa-dba0-4000-af84-950ccf925be4",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "y_train_num = label_encoder.fit_transform(y_train)\n",
    "y_test_num = label_encoder.fit_transform(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "344a9f76-3caf-4784-9ce7-10b89ff6b869",
   "metadata": {},
   "outputs": [],
   "source": [
    "decision_tree = DecisionTreeClassifier(random_state=42)\n",
    "# Inicijaliziram model stabla odlučivanja"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4733adae-364f-42e8-bff2-3ac976d992e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier(random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "DecisionTreeClassifier(random_state=42)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "decision_tree.fit(X_train, y_train)  # Treniranje modela"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "979b1815-ca93-455c-9006-660acf1570c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predict_dt = decision_tree.predict(X_test)  # Predviđanja modela za testne podatke"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "40dc2537-aa31-42c0-b9d0-4239408fc2db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Accuracy using 10-fold Cross-Validation (Decision Tree): 0.8200000000000001\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "# Initialize your model (Decision Tree Classifier in this case)\n",
    "dt_classifier = DecisionTreeClassifier(random_state=42)\n",
    "\n",
    "# Perform 10-fold cross-validation\n",
    "cv_scores_dt = cross_val_score(dt_classifier, X, y, cv=10)\n",
    "\n",
    "# Calculate the mean accuracy of the cross-validation scores\n",
    "mean_accuracy_dt = cv_scores_dt.mean()\n",
    "\n",
    "print(f\"Mean Accuracy using 10-fold Cross-Validation (Decision Tree): {mean_accuracy_dt}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "656e970d-d6f9-4812-bed4-da9d0afa92c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Matrica zabune:\n",
      "[[ 8  4]\n",
      " [ 3 15]]\n"
     ]
    }
   ],
   "source": [
    "cm_dt = confusion_matrix(y_test, y_predict_dt)\n",
    "print(\"Matrica zabune:\")\n",
    "print(cm_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c57d080e-c6d5-4c79-9ffa-c33404508944",
   "metadata": {},
   "outputs": [],
   "source": [
    "disp_dt = ConfusionMatrixDisplay(confusion_matrix = cm_dt, display_labels=decision_tree.classes_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "619000db-b865-47f1-b267-7bc600196538",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x182aa305c30>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp_dt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4c92b896-9860-45cf-a54d-1267b045464a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision :  0.7666666666666667\n",
      "Recall :  0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "precision_dt = precision_score(y_test,y_predict_dt,average='micro')\n",
    "recall_dt = recall_score(y_test, y_predict_dt,average='micro')\n",
    "print(\"Precision : \",precision_dt)\n",
    "print(\"Recall : \",recall_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d5ed34e6-c2e9-4637-acb5-be064caab235",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "accuracy_dt = decision_tree.score(X_test, y_test)\n",
    "print(f\"Accuracy: {accuracy_dt}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d33391e4-23ce-4261-9b60-30a6740d8aa3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1_score:  0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "f1_dt = f1_score(y_test,y_predict_dt,average='micro')\n",
    "print(\"F1_score: \", f1_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "796c3c28-c9ac-4cd7-894a-6d9698d0461b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Demented       0.73      0.67      0.70        12\n",
      " Nondemented       0.79      0.83      0.81        18\n",
      "\n",
      "    accuracy                           0.77        30\n",
      "   macro avg       0.76      0.75      0.75        30\n",
      "weighted avg       0.76      0.77      0.76        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_predict_dt))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c96ba83b-b904-4ab5-a09b-7598f37f0495",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Demented', 'Nondemented'], dtype=object)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "decision_tree.classes_"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05523d83-e0aa-4d5b-bb2c-d5b13030a8bc",
   "metadata": {},
   "source": [
    "ROC (Receiver Operating Characteristic) curve i AUC (Area Under the Curve) krivulje"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e508a0aa-94ce-4741-b923-beaca701fd5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test_num = label_encoder.fit_transform(y_test)\n",
    "#pretvaramo u numericke vrijednosti: 0 i 1 umjesto: 'Demented', 'Nondemented'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "73cb565b-f091-438b-a083-1099f975afbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probs_dt = decision_tree.predict_proba(X_test)[:, 1] #dobiti vjerojatnosti samo za pozitivnu klasu."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f53e7050-35a4-4007-b8bf-52c6d8c57aae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7500000000000002"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test_num, y_probs_dt)\n",
    "auc = roc_auc_score(y_test_num, y_probs_dt)\n",
    "auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a601c037-95f6-423e-8313-779548079f27",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, label=f'AUC = {auc:.2f}')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', color='r', label='Random')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b27481b6-48eb-471a-85d4-0908510cfde0",
   "metadata": {},
   "source": [
    "2.Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd829131-845a-409e-8a27-ebbf7f3355e3",
   "metadata": {},
   "source": [
    "Model slučajne šume performira bolje od stabla odluke jer izbjegava problem prenaučenosti. Modeli temeljeni na slučajnim šumama sastoje se od različitih stabala odluke, svako malo drugačije od ostalih. Koristeći algoritam većinskog glasovanja, ansambl čini predviđanja na temelju svakog pojedinačnog modela stabla odluke (bagging). Kao rezultat, količina prenaučenosti se smanjuje dok se održava prediktivna sposobnost svakog stabla."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "40912d37-0e13-4088-a4f1-511042fd239b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "y_train_num = label_encoder.fit_transform(y_train)\n",
    "y_test_num = label_encoder.fit_transform(y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "958ac9d0-fc8c-447a-aa30-a3658065a87c",
   "metadata": {},
   "source": [
    "\n",
    "# Define hyperparameters to search through\n",
    "param_grid = {\n",
    "    'n_estimators': [50, 100, 150],\n",
    "    'max_depth': [None, 5, 10],\n",
    "    'min_samples_leaf': [1, 2, 4],\n",
    "     'max_features': ['sqrt', 'log2', None]\n",
    "}\n",
    "\n",
    "# Create the GridSearchCV object\n",
    "grid_search = GridSearchCV(RandomForestClassifier(random_state=42), param_grid, cv=5, scoring='accuracy')\n",
    "\n",
    "# Fit the grid search to your data\n",
    "grid_search.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "id": "96e6bba0-f7a7-4198-a57c-bb71be854993",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-20 {color: black;}#sk-container-id-20 pre{padding: 0;}#sk-container-id-20 div.sk-toggleable {background-color: white;}#sk-container-id-20 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-20 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-20 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-20 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-20 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-20 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-20 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-20 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-20 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-20 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-20 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-20 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-20 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-20 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-20 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-20 div.sk-item {position: relative;z-index: 1;}#sk-container-id-20 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-20 div.sk-item::before, #sk-container-id-20 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-20 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-20 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-20 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-20 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-20 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-20 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-20 div.sk-label-container {text-align: center;}#sk-container-id-20 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-20 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-20\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(min_samples_leaf=2, random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-54\" type=\"checkbox\" checked><label for=\"sk-estimator-id-54\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(min_samples_leaf=2, random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(min_samples_leaf=2, random_state=42)"
      ]
     },
     "execution_count": 235,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_classifier = RandomForestClassifier(\n",
    "    n_estimators=100,  # Number of trees\n",
    "    max_depth=None,  # Max depth of trees\n",
    "    min_samples_leaf=2,  # Minimum samples per leaf\n",
    "    max_features='sqrt',  # Number of features to consider at each split\n",
    "    random_state=42  # Setting a random state for reproducibility\n",
    ")\n",
    "\n",
    "# Fit the model on your training data\n",
    "rf_classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "id": "aacc10d4-a3b5-4aca-a7c6-857709287c20",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_rf = rf_classifier.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "id": "b8a042d2-be35-4521-8058-26023b891c58",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8\n"
     ]
    }
   ],
   "source": [
    "accuracy_rf = rf_classifier.score(X_test, y_test)\n",
    "print(f\"Accuracy: {accuracy_rf}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "id": "563e8351-63b6-4832-a90d-03d637a720d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1_score:  0.8000000000000002\n",
      "Precision :  0.8\n",
      "Recall :  0.8\n"
     ]
    }
   ],
   "source": [
    "f1_rf = f1_score(y_test,predictions_rf,average='micro')\n",
    "precision_rf = precision_score(y_test,predictions_rf,average='micro')\n",
    "recall_rf = recall_score(y_test, predictions_rf,average='micro')\n",
    "print(\"F1_score: \",f1_rf)\n",
    "print(\"Precision : \",precision_rf)\n",
    "print(\"Recall : \",recall_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "id": "fe7d863e-7e8d-4d7b-9c4a-b06ee7c77be7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Demented       0.80      0.67      0.73        12\n",
      " Nondemented       0.80      0.89      0.84        18\n",
      "\n",
      "    accuracy                           0.80        30\n",
      "   macro avg       0.80      0.78      0.78        30\n",
      "weighted avg       0.80      0.80      0.80        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,predictions_rf))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 240,
   "id": "94040605-a974-48f9-9785-243fae8949c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Matrica zabune:\n",
      "[[ 8  4]\n",
      " [ 2 16]]\n"
     ]
    }
   ],
   "source": [
    "cm_rf = confusion_matrix(y_test, predictions_rf)\n",
    "print(\"Matrica zabune:\")\n",
    "print(cm_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "id": "97e6eff3-33b1-406f-88ca-bbd5d9e74856",
   "metadata": {},
   "outputs": [],
   "source": [
    "disp_rf = ConfusionMatrixDisplay(confusion_matrix = cm_rf, display_labels=rf_classifier.classes_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "id": "2445e9db-b3d6-4a78-86ec-446ca35fbb00",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x182d213dde0>"
      ]
     },
     "execution_count": 242,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp_rf.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49c924ac-6905-4cfe-aabf-6b7506410002",
   "metadata": {},
   "source": [
    "ROC (Receiver Operating Characteristic) curve i AUC (Area Under the Curve) krivulje"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ab63da3-bbe4-43fa-abb0-9c06ce99025c",
   "metadata": {},
   "source": [
    "ROC krivulje\n",
    "Grafički prikaz izvedbe binarnog klasifikatora kao funkcija različitih pragova\r\n",
    "korištenih pri klasifikaciji"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd806ac3-6559-448e-9073-d37fe2d091d3",
   "metadata": {},
   "source": [
    "Razina ispravno pozitivnih (TPR): To je omjer točno predviđenih pozitivnih primjera u odnosu na sve stvarne pozitivne primjere. \r\n",
    "\r\n",
    "Razina lažno pozitivnih (FPR): To je omjer neispravno predviđenih pozitivnih primjera u odnosu na sve stvarne negativne primjer\n",
    "\n",
    "Veći AUC – bolji model.\n",
    "\r\n",
    "\r\n",
    "\r\n",
    "\r\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "id": "4738460e-55aa-4dcf-ae00-f5ad4edfc182",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probs_rf = rf_classifier.predict_proba(X_test)[:, 1] #dobiti vjerojatnosti samo za pozitivnu klasu."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "id": "838dd746-98d5-4b48-858c-431368991882",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8101851851851852"
      ]
     },
     "execution_count": 244,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test_num, y_probs_rf)\n",
    "auc = roc_auc_score(y_test_num, y_probs_rf)\n",
    "auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 245,
   "id": "e713d2b6-1be5-447b-9d54-e36f8ab3597e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, label=f'AUC = {auc:.2f}')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', color='r', label='Random')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f0a036b-f158-43e2-a8c5-705949b5b862",
   "metadata": {},
   "source": [
    "3.Support Vector Machine (SVM)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efaa33bc-5855-4f8d-ac2c-b33bc8b2c804",
   "metadata": {},
   "source": [
    "Ova metoda uključuje određivanje klase podataka pomoću odgovarajućih hiper-ravnina u višedimenzionalnom prostoru. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 246,
   "id": "0152ea7c-680d-4057-8688-232b80e67b8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "y_train_num = label_encoder.fit_transform(y_train)\n",
    "y_test_num = label_encoder.fit_transform(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 247,
   "id": "5c5705b4-4fda-4779-bb33-b7f3a92d375e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-21 {color: black;}#sk-container-id-21 pre{padding: 0;}#sk-container-id-21 div.sk-toggleable {background-color: white;}#sk-container-id-21 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-21 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-21 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-21 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-21 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-21 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-21 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-21 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-21 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-21 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-21 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-21 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-21 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-21 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-21 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-21 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-21 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-21 div.sk-item {position: relative;z-index: 1;}#sk-container-id-21 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-21 div.sk-item::before, #sk-container-id-21 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-21 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-21 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-21 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-21 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-21 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-21 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-21 div.sk-label-container {text-align: center;}#sk-container-id-21 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-21 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-21\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SVC(kernel=&#x27;linear&#x27;, probability=True)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-55\" type=\"checkbox\" checked><label for=\"sk-estimator-id-55\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC(kernel=&#x27;linear&#x27;, probability=True)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "SVC(kernel='linear', probability=True)"
      ]
     },
     "execution_count": 247,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_classifier = SVC(kernel='linear', probability=True)  # You can specify the kernel type ('linear', 'rbf', 'poly', etc.)\n",
    "svm_classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 248,
   "id": "5f7991dd-7ca5-403f-a42e-9b74e14886d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8666666666666667\n"
     ]
    }
   ],
   "source": [
    "# Making predictions on the test set using the best estimator\n",
    "y_predict_svm = svm_classifier.predict(X_test)\n",
    "\n",
    "# Calculating accuracy\n",
    "accuracy_svm = accuracy_score(y_test, y_predict_svm)\n",
    "print(f\"Accuracy: {accuracy_svm}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 249,
   "id": "f275ec63-baad-4278-a878-660c2f45b3ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Matrica zabune:\n",
      "[[ 8  4]\n",
      " [ 0 18]]\n"
     ]
    }
   ],
   "source": [
    "cm_svm = confusion_matrix(y_test, y_predict_svm)\n",
    "print(\"Matrica zabune:\")\n",
    "print(cm_svm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 250,
   "id": "f1e266fb-1986-438e-9e48-7a4e18405d5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "disp_svm = ConfusionMatrixDisplay(confusion_matrix = cm_svm, display_labels=svm_classifier.classes_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "id": "59816c53-e4e0-42bd-ab32-2b5d442a5499",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x182d178ec50>"
      ]
     },
     "execution_count": 251,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp_svm.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 252,
   "id": "8f9c8ce8-25ab-4f91-a678-d711d6de72fe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1_score:  0.8666666666666667\n",
      "Precision :  0.8666666666666667\n",
      "Recall :  0.8666666666666667\n"
     ]
    }
   ],
   "source": [
    "f1_svm = f1_score(y_test,y_predict_svm,average='micro')\n",
    "print(\"F1_score: \",f1_svm)\n",
    "precision_svm = precision_score(y_test, y_predict_svm,average='micro')\n",
    "recall_svm = recall_score(y_test,y_predict_svm,average='micro')\n",
    "print(\"Precision : \",precision_svm)\n",
    "print(\"Recall : \",recall_svm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 253,
   "id": "7ea265fd-7631-4199-a207-3e48863faad9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Demented       1.00      0.67      0.80        12\n",
      " Nondemented       0.82      1.00      0.90        18\n",
      "\n",
      "    accuracy                           0.87        30\n",
      "   macro avg       0.91      0.83      0.85        30\n",
      "weighted avg       0.89      0.87      0.86        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_predict_svm,  zero_division=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4951ab98-d700-4806-8ff9-8d46df41c6d1",
   "metadata": {},
   "source": [
    "ROC (Receiver Operating Characteristic) curve i AUC (Area Under the Curve) krivulje"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "id": "5f1a0022-e34d-42a1-b79c-79605a9f7635",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probs_svm = svm_classifier.predict_proba(X_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "id": "6842bf15-6b16-470a-bb5e-b02bc28b20f2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.875"
      ]
     },
     "execution_count": 255,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test_num, y_probs_svm)\n",
    "auc = roc_auc_score(y_test_num, y_probs_svm)\n",
    "auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "id": "7a409685-b40b-4aa1-aa0a-8c291e223db3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, label=f'AUC = {auc:.2f}')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', color='r', label='Random')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ffcd34fb-9e70-4fc3-be7d-1c0781e8055a",
   "metadata": {},
   "source": [
    "4. XGBoost"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2351829-8a4d-4640-8a8e-72df689392f2",
   "metadata": {},
   "source": [
    " Proces implementacije gradijentno pojačanih stabala odlučivanja za maksimalnu brzinu i performanse."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "id": "060fedd7-0f81-4c10-ba5b-4670d8554de5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-22 {color: black;}#sk-container-id-22 pre{padding: 0;}#sk-container-id-22 div.sk-toggleable {background-color: white;}#sk-container-id-22 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-22 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-22 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-22 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-22 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-22 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-22 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-22 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-22 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-22 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-22 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-22 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-22 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-22 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-22 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-22 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-22 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-22 div.sk-item {position: relative;z-index: 1;}#sk-container-id-22 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-22 div.sk-item::before, #sk-container-id-22 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-22 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-22 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-22 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-22 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-22 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-22 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-22 div.sk-label-container {text-align: center;}#sk-container-id-22 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-22 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-22\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=42, ...)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-56\" type=\"checkbox\" checked><label for=\"sk-estimator-id-56\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">XGBClassifier</label><div class=\"sk-toggleable__content\"><pre>XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=42, ...)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
       "              colsample_bylevel=None, colsample_bynode=None,\n",
       "              colsample_bytree=None, device=None, early_stopping_rounds=None,\n",
       "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "              gamma=None, grow_policy=None, importance_type=None,\n",
       "              interaction_constraints=None, learning_rate=None, max_bin=None,\n",
       "              max_cat_threshold=None, max_cat_to_onehot=None,\n",
       "              max_delta_step=None, max_depth=None, max_leaves=None,\n",
       "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
       "              multi_strategy=None, n_estimators=None, n_jobs=None,\n",
       "              num_parallel_tree=None, random_state=42, ...)"
      ]
     },
     "execution_count": 257,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_classifier = XGBClassifier(random_state=42)\n",
    "xgb_classifier.fit(X_train, y_train_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "id": "a35ebe57-e202-4b25-876a-17bf1e6e31c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_predict_xgb = xgb_classifier.predict(X_test)  # Predviđanja modela za testne podatke"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "id": "74366b2f-772d-4977-90e9-4d3301b4cd13",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Matrica zabune:\n",
      "[[ 8  4]\n",
      " [ 3 15]]\n"
     ]
    }
   ],
   "source": [
    "cm_xgb = confusion_matrix(y_test_num, y_predict_xgb)\n",
    "print(\"Matrica zabune:\")\n",
    "print(cm_xgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "id": "423dce7a-95b3-4a77-b759-7c9854aa0682",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x182d2526530>"
      ]
     },
     "execution_count": 260,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp_xgb = ConfusionMatrixDisplay(confusion_matrix = cm_xgb, display_labels=xgb_classifier.classes_)\n",
    "disp_xgb.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "id": "eeffe991-df7f-463a-8bcc-0479bf6cf4f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1_score:  0.7666666666666667\n",
      "Precision :  0.7666666666666667\n",
      "Recall :  0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "f1_xgb = f1_score(y_test_num,y_predict_xgb,average='micro')\n",
    "print(\"F1_score: \",f1_xgb)\n",
    "precision_xgb = precision_score(y_test_num, y_predict_xgb,average='micro')\n",
    "recall_xgb = recall_score(y_test_num,y_predict_xgb,average='micro')\n",
    "print(\"Precision : \",precision_xgb)\n",
    "print(\"Recall : \",recall_xgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "id": "e8062fbf-908e-4e41-8105-0f2efb8d305d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "accuracy_xgb = xgb_classifier.score(X_test, y_test_num)\n",
    "print(f\"Accuracy: {accuracy_xgb}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 263,
   "id": "f3841d9c-9b4d-434a-9a14-da3f5dca7773",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.73      0.67      0.70        12\n",
      "           1       0.79      0.83      0.81        18\n",
      "\n",
      "    accuracy                           0.77        30\n",
      "   macro avg       0.76      0.75      0.75        30\n",
      "weighted avg       0.76      0.77      0.76        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test_num,y_predict_xgb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "id": "3259f02e-cf6a-4844-b0ca-12b89b981c0c",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probs_xgb = xgb_classifier.predict_proba(X_test)[:, 1] #dobiti vjerojatnosti samo za pozitivnu klasu."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 265,
   "id": "e52441a3-3894-4f29-bd8b-22cb42724296",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7453703703703705"
      ]
     },
     "execution_count": 265,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test_num, y_probs_xgb)\n",
    "auc = roc_auc_score(y_test_num, y_probs_xgb)\n",
    "auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 266,
   "id": "6575ff0d-529a-4ee8-9197-5e27b8c3ce46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, label=f'AUC = {auc:.2f}')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', color='r', label='Random')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5f0e99a-78d2-4be7-b147-397c1dea262c",
   "metadata": {},
   "source": [
    "5. Voting"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3cc3017-6388-46f5-89ae-849e7173df78",
   "metadata": {},
   "source": [
    "Jedan od najjednostavnijih načina kombiniranja predviđanja iz više algoritama za učenje. Glasajući klasifikatori zapravo nisu klasifikatori, već više podsjećaju na omotače (wrappere) za više klasifikatora koji su trenirani i evaluirani istovremeno kako bi se iskoristile njihove specifične karakteristike. \n",
    "\n",
    "Hard voting: Klasa s najviše glasova bit će odabrana u ovom slučaju.\r\n",
    " (Soft votig): Ovo uključuje zbrajanje vektora vjerojatnosti za svaku predviđenu klasu (za sve klasifikatore) i odabir one koja predstavlja najvišu vrijednoi)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "id": "35d15a91-c5b5-4be7-b6a9-7cb85a946904",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "y_train_num = label_encoder.fit_transform(y_train)\n",
    "y_test_num = label_encoder.fit_transform(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "id": "7c6e1496-d54f-496f-93db-b6224daa3749",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Initialize individual classifiers\n",
    "logistic_clf = LogisticRegression(max_iter=1000, random_state=42)\n",
    "tree_clf = DecisionTreeClassifier(random_state=42)\n",
    "svm_clf = SVC(probability=True, random_state=42)\n",
    "rf_clf = RandomForestClassifier(n_estimators=100, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "id": "4bde19bb-ee31-468a-b85f-5fc1aae4707b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a voting classifier (both hard and soft voting)\n",
    "voting_clf_hard = VotingClassifier(\n",
    "    estimators=[('lr', logistic_clf), ('dt', tree_clf), ('svm', svm_clf), ('rf', rf_clf)],\n",
    "    voting='hard'\n",
    ")\n",
    "\n",
    "voting_clf_soft = VotingClassifier(\n",
    "    estimators=[('lr', logistic_clf), ('dt', tree_clf), ('svm', svm_clf), ('rf', rf_clf)],\n",
    "    voting='soft'\n",
    ")\n",
    "logistic_classifier = LogisticRegression(random_state=42, max_iter=1000)\n",
    "\n",
    "gb_classifier = GradientBoostingClassifier(random_state=42)\n",
    "voting_classifier = VotingClassifier(\n",
    "    estimators=[('rf', rf_classifier), ('svm', svm_classifier), ('logistic', logistic_classifier), ('gb', gb_classifier)],\n",
    "    voting='soft'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "id": "72180174-81d9-4b99-80e9-55bd04f42c8d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-29 {color: black;}#sk-container-id-29 pre{padding: 0;}#sk-container-id-29 div.sk-toggleable {background-color: white;}#sk-container-id-29 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-29 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-29 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-29 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-29 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-29 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-29 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-29 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-29 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-29 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-29 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-29 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-29 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-29 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-29 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-29 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-29 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-29 div.sk-item {position: relative;z-index: 1;}#sk-container-id-29 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-29 div.sk-item::before, #sk-container-id-29 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-29 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-29 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-29 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-29 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-29 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-29 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-29 div.sk-label-container {text-align: center;}#sk-container-id-29 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-29 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-29\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>VotingClassifier(estimators=[(&#x27;lr&#x27;,\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             (&#x27;dt&#x27;, DecisionTreeClassifier(random_state=42)),\n",
       "                             (&#x27;svm&#x27;, SVC(probability=True, random_state=42)),\n",
       "                             (&#x27;rf&#x27;, RandomForestClassifier(random_state=42))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-87\" type=\"checkbox\" ><label for=\"sk-estimator-id-87\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">VotingClassifier</label><div class=\"sk-toggleable__content\"><pre>VotingClassifier(estimators=[(&#x27;lr&#x27;,\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             (&#x27;dt&#x27;, DecisionTreeClassifier(random_state=42)),\n",
       "                             (&#x27;svm&#x27;, SVC(probability=True, random_state=42)),\n",
       "                             (&#x27;rf&#x27;, RandomForestClassifier(random_state=42))])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lr</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-88\" type=\"checkbox\" ><label for=\"sk-estimator-id-88\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(max_iter=1000, random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>dt</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-89\" type=\"checkbox\" ><label for=\"sk-estimator-id-89\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>svm</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-90\" type=\"checkbox\" ><label for=\"sk-estimator-id-90\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC(probability=True, random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>rf</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-91\" type=\"checkbox\" ><label for=\"sk-estimator-id-91\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=42)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "VotingClassifier(estimators=[('lr',\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             ('dt', DecisionTreeClassifier(random_state=42)),\n",
       "                             ('svm', SVC(probability=True, random_state=42)),\n",
       "                             ('rf', RandomForestClassifier(random_state=42))])"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "voting_clf_hard.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "id": "f11bec92-a9e9-4acc-9782-25644b094e03",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-30 {color: black;}#sk-container-id-30 pre{padding: 0;}#sk-container-id-30 div.sk-toggleable {background-color: white;}#sk-container-id-30 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-30 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-30 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-30 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-30 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-30 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-30 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-30 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-30 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-30 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-30 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-30 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-30 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-30 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-30 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-30 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-30 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-30 div.sk-item {position: relative;z-index: 1;}#sk-container-id-30 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-30 div.sk-item::before, #sk-container-id-30 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-30 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-30 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-30 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-30 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-30 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-30 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-30 div.sk-label-container {text-align: center;}#sk-container-id-30 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-30 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-30\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>VotingClassifier(estimators=[(&#x27;lr&#x27;,\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             (&#x27;dt&#x27;, DecisionTreeClassifier(random_state=42)),\n",
       "                             (&#x27;svm&#x27;, SVC(probability=True, random_state=42)),\n",
       "                             (&#x27;rf&#x27;, RandomForestClassifier(random_state=42))],\n",
       "                 voting=&#x27;soft&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-92\" type=\"checkbox\" ><label for=\"sk-estimator-id-92\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">VotingClassifier</label><div class=\"sk-toggleable__content\"><pre>VotingClassifier(estimators=[(&#x27;lr&#x27;,\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             (&#x27;dt&#x27;, DecisionTreeClassifier(random_state=42)),\n",
       "                             (&#x27;svm&#x27;, SVC(probability=True, random_state=42)),\n",
       "                             (&#x27;rf&#x27;, RandomForestClassifier(random_state=42))],\n",
       "                 voting=&#x27;soft&#x27;)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>lr</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-93\" type=\"checkbox\" ><label for=\"sk-estimator-id-93\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(max_iter=1000, random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>dt</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-94\" type=\"checkbox\" ><label for=\"sk-estimator-id-94\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier(random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>svm</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-95\" type=\"checkbox\" ><label for=\"sk-estimator-id-95\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SVC</label><div class=\"sk-toggleable__content\"><pre>SVC(probability=True, random_state=42)</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><label>rf</label></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-96\" type=\"checkbox\" ><label for=\"sk-estimator-id-96\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=42)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "VotingClassifier(estimators=[('lr',\n",
       "                              LogisticRegression(max_iter=1000,\n",
       "                                                 random_state=42)),\n",
       "                             ('dt', DecisionTreeClassifier(random_state=42)),\n",
       "                             ('svm', SVC(probability=True, random_state=42)),\n",
       "                             ('rf', RandomForestClassifier(random_state=42))],\n",
       "                 voting='soft')"
      ]
     },
     "execution_count": 309,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "voting_clf_soft.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "id": "68ff7e0b-6a45-49c9-b045-4189497e067d",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_hard = voting_clf_hard.predict(X_test)\n",
    "y_pred_soft = voting_clf_soft.predict(X_test)\n",
    "\n",
    "voting_classifier.fit(X_train, y_train)\n",
    "y_pred_voting = voting_classifier.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 316,
   "id": "693915e4-8768-40b9-a789-54838d71ad5a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy (Hard Voting): 0.7333333333333333\n",
      "Accuracy voting: 0.8666666666666667\n"
     ]
    }
   ],
   "source": [
    "accuracy_hard = accuracy_score(y_test, y_pred_hard)\n",
    "print(f'Accuracy (Hard Voting): {accuracy_hard}')\n",
    "accuracy_voting = accuracy_score(y_test, y_pred_voting)\n",
    "print(f'Accuracy voting: {accuracy_voting}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "id": "15fa97a0-cf2e-40fd-bcd8-04bc80f97bd5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy (Soft Voting): 0.7666666666666667\n"
     ]
    }
   ],
   "source": [
    "accuracy_soft = accuracy_score(y_test, y_pred_soft)\n",
    "print(f'Accuracy (Soft Voting): {accuracy_soft}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "id": "a6b710d1-f211-4818-82ff-dde2b295af3a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "F1_score:  0.8666666666666667\n",
      "Precision :  0.8666666666666667\n",
      "Recall :  0.8666666666666667\n"
     ]
    }
   ],
   "source": [
    "f1_soft = f1_score(y_test,y_pred_voting,average='micro')\n",
    "print(\"F1_score: \",f1_soft)\n",
    "precision_soft = precision_score(y_test, y_pred_voting,average='micro')\n",
    "recall_soft = recall_score(y_test,y_pred_voting,average='micro')\n",
    "print(\"Precision : \",precision_soft)\n",
    "print(\"Recall : \",recall_soft)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "id": "1dc52014-2880-4dcc-8468-f4bdc64c7826",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Matrica zabune:\n",
      "[[ 8  4]\n",
      " [ 0 18]]\n"
     ]
    }
   ],
   "source": [
    "conf_matrix_soft = confusion_matrix(y_test, y_pred_voting)\n",
    "print(\"Matrica zabune:\")\n",
    "print(conf_matrix_soft)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 320,
   "id": "432fbb0c-3e77-4fde-a30b-21519ffc47da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x182d1f3f580>"
      ]
     },
     "execution_count": 320,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp_soft = ConfusionMatrixDisplay(confusion_matrix = conf_matrix_soft, display_labels=voting_clf_soft.classes_)\n",
    "disp_soft.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 321,
   "id": "a3d12788-52bc-4385-a43e-1843798db8cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "    Demented       0.73      0.67      0.70        12\n",
      " Nondemented       0.79      0.83      0.81        18\n",
      "\n",
      "    accuracy                           0.77        30\n",
      "   macro avg       0.76      0.75      0.75        30\n",
      "weighted avg       0.76      0.77      0.76        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,y_pred_soft))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d560ad63-a12e-4071-863c-6d47442a9283",
   "metadata": {},
   "source": [
    "ROC (Receiver Operating Characteristic) curve i AUC (Area Under the Curve) krivulje"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 322,
   "id": "d8a31b60-716a-4877-b4ba-90c70fa1da9b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_probs_soft = voting_clf_soft.predict_proba(X_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 323,
   "id": "ec32b36f-b3d4-445e-906c-9260395c0f26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7777777777777779"
      ]
     },
     "execution_count": 323,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test_num, y_probs_soft)\n",
    "auc = roc_auc_score(y_test_num, y_probs_soft)\n",
    "auc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 324,
   "id": "3408beeb-2940-49a2-bf6c-433760486ebf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, label=f'AUC = {auc:.2f}')\n",
    "plt.plot([0, 1], [0, 1], linestyle='--', color='r', label='Random')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c87f1baf-8416-4fb3-8057-8cc9fe7bb836",
   "metadata": {},
   "source": [
    "Comparison of accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 325,
   "id": "a1b5fe10-961a-4347-a1b5-429fb8f51391",
   "metadata": {},
   "outputs": [],
   "source": [
    "classifiers = [\"Decision Tree\", \"Random Forest\", \"SVM\", \"XGBoost\", \"Voting\"]\n",
    "accuracies = [accuracy_dt, accuracy_rf, accuracy_svm, accuracy_xgb, accuracy_soft]  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 326,
   "id": "04f70878-f43d-42b8-8d75-d4480167f62c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(classifiers, accuracies, color='skyblue')\n",
    "plt.xlabel('Classifiers')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Accuracy Comparison of Classifiers')\n",
    "plt.ylim(0, 1)  # Set the y-axis limit for better visualization\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 327,
   "id": "74ff9049-ffc8-4005-a197-887ed1129786",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.7666666666666667,\n",
       " 0.8,\n",
       " 0.8666666666666667,\n",
       " 0.7666666666666667,\n",
       " 0.8666666666666667]"
      ]
     },
     "execution_count": 327,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifiers = [\"Decision Tree\", \"Random Forest\", \"SVM\", \"XGBoost\", \"Voting\"]\n",
    "precision_all = [precision_dt, precision_rf, precision_svm, precision_xgb, precision_soft] \n",
    "precision_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 328,
   "id": "4704e5c9-d90c-4ded-bb6f-08b89423b538",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(classifiers, precision_all, color='skyblue')\n",
    "plt.xlabel('Classifiers')\n",
    "plt.ylabel('Precision')\n",
    "plt.title('Precision Comparison of Classifiers')\n",
    "plt.ylim(0.7, 0.9)  # Set the y-axis limit for better visualization\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 329,
   "id": "be2cb8bb-56a1-4677-8a55-8b8400f29787",
   "metadata": {},
   "outputs": [],
   "source": [
    "classifiers = [\"Decision Tree\", \"Random Forest\", \"SVM\", \"XGBoost\", \"Voting\"]\n",
    "recall_all = [recall_dt, recall_rf, recall_svm, recall_xgb,recall_soft] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 330,
   "id": "58181d65-9e16-485f-af82-cdef8680f75d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(classifiers,recall_all, color='skyblue')\n",
    "plt.xlabel('Classifiers')\n",
    "plt.ylabel('Recall')\n",
    "plt.title('Recall Comparison of Classifiers')\n",
    "plt.ylim(0.7, 0.9)  # Set the y-axis limit for better visualization\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "id": "a46a8471-8026-4890-9033-e59621610d86",
   "metadata": {},
   "outputs": [],
   "source": [
    "classifiers = [\"Decision Tree\", \"Random Forest\", \"SVM\", \"XGBoost\", \"Voting\"]\n",
    "f1_all = [f1_dt, f1_rf, f1_svm, f1_xgb,f1_soft] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 332,
   "id": "8e7b04a5-94b0-4642-a50f-ccff200966fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(classifiers,f1_all, color='skyblue')\n",
    "plt.xlabel('Classifiers')\n",
    "plt.ylabel('F1')\n",
    "plt.title('F1 Comparison of Classifiers')\n",
    "plt.ylim(0.7, 0.9)  # Set the y-axis limit for better visualization\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a52c3fca-186d-4f5c-9a64-0cd5b2f77c8d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eac1fe8d-60fa-457a-ad84-cc71f4168952",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "My Python Environment",
   "language": "python",
   "name": "myenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}