{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "exposed-browse",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt \n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "import xgboost as xgb\n",
    "import torch\n",
    "\n",
    "def eval(model,x_test,y_test, nn = None):\n",
    "    if nn == None:\n",
    "        print(classification_report(y_test,model.predict(x_test)))\n",
    "    else:\n",
    "        y_pred = model(torch.tensor(x_test.astype(np.float32)))\n",
    "        y_pred  = y_pred.cpu().detach().numpy() \n",
    "        y_pred = (y_pred > 0.5)\n",
    "        y_pred  = np.asarray(y_pred, dtype=np.int32)\n",
    "        print(classification_report(y_test,y_pred))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "unlimited-modern",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('water_potability.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "simplified-ranch",
   "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>ph</th>\n",
       "      <th>Hardness</th>\n",
       "      <th>Solids</th>\n",
       "      <th>Chloramines</th>\n",
       "      <th>Sulfate</th>\n",
       "      <th>Conductivity</th>\n",
       "      <th>Organic_carbon</th>\n",
       "      <th>Trihalomethanes</th>\n",
       "      <th>Turbidity</th>\n",
       "      <th>Potability</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2785.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>2495.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>3114.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "      <td>3276.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.080795</td>\n",
       "      <td>196.369496</td>\n",
       "      <td>22014.092526</td>\n",
       "      <td>7.122277</td>\n",
       "      <td>333.775777</td>\n",
       "      <td>426.205111</td>\n",
       "      <td>14.284970</td>\n",
       "      <td>66.396293</td>\n",
       "      <td>3.966786</td>\n",
       "      <td>0.390110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.594320</td>\n",
       "      <td>32.879761</td>\n",
       "      <td>8768.570828</td>\n",
       "      <td>1.583085</td>\n",
       "      <td>41.416840</td>\n",
       "      <td>80.824064</td>\n",
       "      <td>3.308162</td>\n",
       "      <td>16.175008</td>\n",
       "      <td>0.780382</td>\n",
       "      <td>0.487849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>47.432000</td>\n",
       "      <td>320.942611</td>\n",
       "      <td>0.352000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>181.483754</td>\n",
       "      <td>2.200000</td>\n",
       "      <td>0.738000</td>\n",
       "      <td>1.450000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.093092</td>\n",
       "      <td>176.850538</td>\n",
       "      <td>15666.690297</td>\n",
       "      <td>6.127421</td>\n",
       "      <td>307.699498</td>\n",
       "      <td>365.734414</td>\n",
       "      <td>12.065801</td>\n",
       "      <td>55.844536</td>\n",
       "      <td>3.439711</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.036752</td>\n",
       "      <td>196.967627</td>\n",
       "      <td>20927.833607</td>\n",
       "      <td>7.130299</td>\n",
       "      <td>333.073546</td>\n",
       "      <td>421.884968</td>\n",
       "      <td>14.218338</td>\n",
       "      <td>66.622485</td>\n",
       "      <td>3.955028</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.062066</td>\n",
       "      <td>216.667456</td>\n",
       "      <td>27332.762127</td>\n",
       "      <td>8.114887</td>\n",
       "      <td>359.950170</td>\n",
       "      <td>481.792304</td>\n",
       "      <td>16.557652</td>\n",
       "      <td>77.337473</td>\n",
       "      <td>4.500320</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>14.000000</td>\n",
       "      <td>323.124000</td>\n",
       "      <td>61227.196008</td>\n",
       "      <td>13.127000</td>\n",
       "      <td>481.030642</td>\n",
       "      <td>753.342620</td>\n",
       "      <td>28.300000</td>\n",
       "      <td>124.000000</td>\n",
       "      <td>6.739000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                ph     Hardness        Solids  Chloramines      Sulfate  \\\n",
       "count  2785.000000  3276.000000   3276.000000  3276.000000  2495.000000   \n",
       "mean      7.080795   196.369496  22014.092526     7.122277   333.775777   \n",
       "std       1.594320    32.879761   8768.570828     1.583085    41.416840   \n",
       "min       0.000000    47.432000    320.942611     0.352000   129.000000   \n",
       "25%       6.093092   176.850538  15666.690297     6.127421   307.699498   \n",
       "50%       7.036752   196.967627  20927.833607     7.130299   333.073546   \n",
       "75%       8.062066   216.667456  27332.762127     8.114887   359.950170   \n",
       "max      14.000000   323.124000  61227.196008    13.127000   481.030642   \n",
       "\n",
       "       Conductivity  Organic_carbon  Trihalomethanes    Turbidity   Potability  \n",
       "count   3276.000000     3276.000000      3114.000000  3276.000000  3276.000000  \n",
       "mean     426.205111       14.284970        66.396293     3.966786     0.390110  \n",
       "std       80.824064        3.308162        16.175008     0.780382     0.487849  \n",
       "min      181.483754        2.200000         0.738000     1.450000     0.000000  \n",
       "25%      365.734414       12.065801        55.844536     3.439711     0.000000  \n",
       "50%      421.884968       14.218338        66.622485     3.955028     0.000000  \n",
       "75%      481.792304       16.557652        77.337473     4.500320     1.000000  \n",
       "max      753.342620       28.300000       124.000000     6.739000     1.000000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "present-wedding",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "informal-allergy",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1224 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,17))\n",
    "matrix = np.triu(data.corr())\n",
    "sns.heatmap(data.corr(), annot=True, linewidth=0, mask=matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "respiratory-assumption",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_feat = data.iloc[:, :-1]\n",
    "data_target = data.iloc[:, -1]\n",
    "st = StandardScaler()\n",
    "data_feat = st.fit_transform(data_feat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fallen-buffer",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(data_feat, data_target, random_state = 42, test_size = 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "accomplished-afternoon",
   "metadata": {},
   "outputs": [],
   "source": [
    "DT = DecisionTreeClassifier(criterion = 'gini').fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "laughing-revelation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.72      0.67      0.69       123\n",
      "           1       0.53      0.59      0.56        79\n",
      "\n",
      "    accuracy                           0.64       202\n",
      "   macro avg       0.63      0.63      0.63       202\n",
      "weighted avg       0.65      0.64      0.64       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(DT,X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "practical-perspective",
   "metadata": {},
   "outputs": [],
   "source": [
    "SV =  SVC(gamma='scale').fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "alive-trick",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.70      0.93      0.80       123\n",
      "           1       0.78      0.39      0.52        79\n",
      "\n",
      "    accuracy                           0.72       202\n",
      "   macro avg       0.74      0.66      0.66       202\n",
      "weighted avg       0.73      0.72      0.69       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(SV,X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "brave-islam",
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb\n",
    "xgb_model = xgb.XGBClassifier(objective=\"binary:logistic\",\n",
    "                              eval_metric='logloss',\n",
    "                              use_label_encoder = False,\n",
    "                              booster='gbtree',reg_lambda=3).fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "controlling-ordinary",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.67      0.72      0.69       123\n",
      "           1       0.50      0.44      0.47        79\n",
      "\n",
      "    accuracy                           0.61       202\n",
      "   macro avg       0.58      0.58      0.58       202\n",
      "weighted avg       0.60      0.61      0.60       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(xgb_model,X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "fluid-responsibility",
   "metadata": {},
   "outputs": [],
   "source": [
    "kn = KNeighborsClassifier(n_neighbors=1,\n",
    "                             algorithm='kd_tree').fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "coordinate-steal",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.67      0.71      0.69       123\n",
      "           1       0.50      0.46      0.48        79\n",
      "\n",
      "    accuracy                           0.61       202\n",
      "   macro avg       0.58      0.58      0.58       202\n",
      "weighted avg       0.60      0.61      0.61       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(kn,X_test,Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "addressed-nothing",
   "metadata": {},
   "outputs": [],
   "source": [
    "class NeuralNetworkModel(torch.nn.Module):\n",
    "\n",
    "    def __init__(self):\n",
    "        super(NeuralNetworkModel, self).__init__()\n",
    "        self.fc1 = torch.nn.Linear(9,2000)\n",
    "        self.fc2 = torch.nn.Linear(2000,1000)\n",
    "        self.fc3 = torch.nn.Linear(1000,1)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.fc1(x)\n",
    "        x = torch.relu(x)\n",
    "        x = self.fc2(x)\n",
    "        x = torch.relu(x)\n",
    "        x = self.fc3(x)\n",
    "        x = torch.sigmoid(x)\n",
    "        return x\n",
    "    \n",
    "model_nn = NeuralNetworkModel()\n",
    "criterion = torch.nn.BCELoss()\n",
    "optimizer = torch.optim.SGD(model_nn.parameters(), lr=0.01)\n",
    " \n",
    "batch_size = 3\n",
    "\n",
    " \n",
    "for epoch in range(6):\n",
    "    loss_score = 0\n",
    "    acc_score = 0\n",
    "    items_total = 0\n",
    "    model_nn.train()\n",
    "    for i in range(0, Y_train.to_numpy().shape[0], batch_size):\n",
    "        X = X_train[i:i+batch_size]\n",
    "        X = torch.tensor(X.astype(np.float32))\n",
    "        Y = Y_train[i:i+batch_size].to_numpy()\n",
    "        Y = torch.tensor(Y.astype(np.float32)).reshape(-1,1)\n",
    " \n",
    "        Y_predictions = model_nn(X)\n",
    "        acc_score += torch.sum((Y_predictions > 0.5) == Y).item()\n",
    "        items_total += Y.shape[0] \n",
    "\n",
    "        optimizer.zero_grad()\n",
    "        loss = criterion(Y_predictions, Y)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "\n",
    "        loss_score += loss.item() * Y.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "brilliant-surge",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.70      0.89      0.78       123\n",
      "           1       0.70      0.41      0.51        79\n",
      "\n",
      "    accuracy                           0.70       202\n",
      "   macro avg       0.70      0.65      0.65       202\n",
      "weighted avg       0.70      0.70      0.68       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(model_nn,X_test,Y_test, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "controlled-measure",
   "metadata": {},
   "outputs": [],
   "source": [
    "class LogisticRegressionModel(torch.nn.Module):\n",
    "\n",
    "    def __init__(self):\n",
    "        super(LogisticRegressionModel, self).__init__()\n",
    "        self.fc = torch.nn.Linear(9,1)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.fc(x)\n",
    "        x = torch.sigmoid(x)\n",
    "        return x\n",
    "    \n",
    "lr_model = LogisticRegressionModel()\n",
    "BATCH_SIZE = 2\n",
    "criterion = torch.nn.BCELoss()\n",
    "optimizer = torch.optim.Adam(lr_model.parameters(), lr = 0.1)\n",
    "loss_score = 0\n",
    "acc_score = 0\n",
    "items_total = 0\n",
    "lr_model.train()\n",
    "for i in range(0, Y_train.shape[0], BATCH_SIZE):\n",
    "    X = X_train[i:i+BATCH_SIZE]\n",
    "    X = torch.tensor(X.astype(np.float32))\n",
    "    Y = Y_train[i:i+BATCH_SIZE].to_numpy()\n",
    "    Y = torch.tensor(Y.astype(np.float32)).reshape(-1,1)\n",
    "    Y_predictions = lr_model(X)\n",
    "    acc_score += torch.sum((Y_predictions > 0.5) == Y).item()\n",
    "    items_total += Y.shape[0] \n",
    "    \n",
    "    optimizer.zero_grad()\n",
    "    loss = criterion(Y_predictions, Y)\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    \n",
    "\n",
    "    loss_score += loss.item() * Y.shape[0] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "environmental-virginia",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.63      0.80      0.71       123\n",
      "           1       0.47      0.28      0.35        79\n",
      "\n",
      "    accuracy                           0.59       202\n",
      "   macro avg       0.55      0.54      0.53       202\n",
      "weighted avg       0.57      0.59      0.57       202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "eval(lr_model,X_test,Y_test, 1)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}