Project_NaiveBayes.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "LtqK8Zyw_-bT"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import csv\n",
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading the Data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "zYpK-J-4tohM",
    "outputId": "3f5c876f-d48c-4a93-eb53-a8a03f6b84da"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Shape: (75000, 1965) , Testing Shape: (25000, 1965)\n"
     ]
    }
   ],
   "source": [
    "# Loading the input data in pandas dataframe\n",
    "x_train = pd.read_csv(\"df_train_2.csv\", header = [0]) \n",
    "x_test = pd.read_csv(\"df_test_2.csv\", header = [0]) \n",
    "print(\"Training Shape:\", x_train.shape,\", Testing Shape:\", x_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "j_Hc9Mlzui8g"
   },
   "outputs": [],
   "source": [
    "# Loading the output data in pandas dataframe\n",
    "y_train = pd.read_csv(\"y_train.csv\",header=None)\n",
    "y_test = pd.read_csv(\"y_test.csv\",header=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "KKd5MVfe5Tcf",
    "outputId": "5590c86c-9860-421a-96e6-05d26d3f77a8"
   },
   "outputs": [],
   "source": [
    "# Converting the inputs and outputs to the numpy array\n",
    "X_train = x_train.copy().to_numpy()\n",
    "X_test = x_test.copy().to_numpy()\n",
    "Y_train = y_train.copy().to_numpy().reshape(-1)\n",
    "Y_test = y_test.copy().to_numpy().reshape(-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mulitnomail NB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------ Results for Mulitnomial NB ------------------\n",
      "alpha = 1e-10 testing: 0.78124 training: 0.84516 \n",
      "alpha = 1e-08 testing: 0.78432 training: 0.8441733333333333 \n",
      "alpha = 1e-06 testing: 0.78844 training: 0.8429466666666666 \n",
      "alpha = 1e-05 testing: 0.79052 training: 0.8417733333333334 \n",
      "alpha = 0.001 testing: 0.79444 training: 0.8382666666666667 \n",
      "alpha = 0.01 testing: 0.7968 training: 0.8358266666666667 \n",
      "alpha = 0.1 testing: 0.79752 training: 0.8319733333333333 \n",
      "alpha = 0.5 testing: 0.79764 training: 0.8284666666666667 \n",
      "alpha = 1 testing: 0.79776 training: 0.8265333333333333 \n",
      "alpha = 5 testing: 0.79792 training: 0.82124 \n",
      "alpha = 10 testing: 0.79784 training: 0.8197466666666666 \n",
      "alpha = 50 testing: 0.79232 training: 0.80588 \n",
      "alpha = 100 testing: 0.7778 training: 0.7900266666666667 \n",
      "alpha = 500 testing: 0.73264 training: 0.7441066666666667 \n",
      "alpha = 1000 testing: 0.7096 training: 0.72024 \n",
      "alpha = 10000 testing: 0.65616 training: 0.66388 \n",
      "alpha = 100000 testing: 0.60672 training: 0.6146133333333333 \n",
      "alpha = 10000000 testing: 0.01656 training: 0.019533333333333333 \n"
     ]
    }
   ],
   "source": [
    "\n",
    "from sklearn.naive_bayes import MultinomialNB \n",
    "\n",
    "# arr stores the values of learning rates to be considered\n",
    "arr = [0.0000000001,0.00000001,0.000001,0.00001,0.001,0.01,0.1,0.5,1,5,10,50,100,500,1000,10000,100000,10000000]\n",
    "\n",
    "train_score = []                                           # for storing train score for different values of learning rate\n",
    "test_score = []                                            # for storing test score for different values of learning rate  \n",
    "\n",
    "\n",
    "\n",
    "print(\"------------------ Results for Mulitnomial NB ------------------\")\n",
    "\n",
    "for i in arr:\n",
    "    \n",
    "    # training the model on X_train \n",
    "    model = MultinomialNB(alpha = i)\n",
    "    model.fit(X_train, Y_train)\n",
    "    \n",
    "    # getting the accuracy on training and testing data\n",
    "    train_score_ = model.score(X_train, Y_train)\n",
    "    test_score_ = model.score(X_test, Y_test)\n",
    "    \n",
    "    # adding the train and test accuracy to train_score and test_score respectively\n",
    "    train_score.append(train_score_)\n",
    "    test_score.append(test_score_)\n",
    "\n",
    "    print(i,\"testing:\",test_score_,\" \",\"training:\",train_score_)\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x19bb03b7550>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "size = 10\n",
    "import matplotlib.pyplot as plt\n",
    "# plotting the loss vs iteration graph for training and validation data\n",
    "figure = plt.figure(1, figsize=(20,10))\n",
    "subplot  = figure.add_subplot(212)\n",
    "# subplot.tick_params(labelrotation=90)\n",
    "subplot.plot(arr[:size], train_score[:size], label = \"Training Accuracy\")\n",
    "subplot.plot(arr[:size], test_score[:size], label = \"Test Accuracy\")\n",
    "# subplot.scatter(X_iteration, Y_train_error)\n",
    "subplot.set_title(\"Accuracy vs Alpha, Model Used : Logistic Regression (L2 Regularization)\")\n",
    "subplot.set_xlabel(\"Alpha\")\n",
    "subplot.set_ylabel(\"Accuracy\")\n",
    "subplot.legend(loc=\"upper right\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:Training\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         1.0       0.63      0.81      0.71       735\n",
      "         2.0       0.70      0.59      0.64       755\n",
      "         3.0       0.91      0.97      0.94       779\n",
      "         4.0       0.89      0.87      0.88       752\n",
      "         5.0       0.96      0.95      0.95       767\n",
      "         6.0       0.76      0.91      0.83       730\n",
      "         7.0       0.90      0.92      0.91       743\n",
      "         8.0       0.98      0.97      0.97       754\n",
      "         9.0       0.98      0.97      0.97       741\n",
      "        10.0       0.73      0.68      0.70       753\n",
      "        11.0       0.78      0.80      0.79       738\n",
      "        12.0       0.76      0.78      0.77       744\n",
      "        13.0       0.95      0.98      0.96       742\n",
      "        14.0       0.77      0.52      0.62       762\n",
      "        15.0       0.63      0.54      0.58       769\n",
      "        16.0       0.47      0.32      0.38       729\n",
      "        17.0       0.80      0.88      0.83       729\n",
      "        18.0       0.83      0.85      0.84       782\n",
      "        19.0       0.74      0.83      0.78       759\n",
      "        20.0       0.58      0.73      0.65       759\n",
      "        21.0       0.76      0.80      0.78       739\n",
      "        22.0       0.99      0.97      0.98       757\n",
      "        23.0       0.92      0.96      0.94       744\n",
      "        24.0       0.80      0.83      0.81       737\n",
      "        25.0       0.96      0.95      0.96       732\n",
      "        26.0       0.63      0.94      0.75       745\n",
      "        27.0       0.91      0.86      0.88       735\n",
      "        28.0       0.72      0.60      0.65       768\n",
      "        29.0       0.84      0.90      0.87       752\n",
      "        30.0       0.90      0.89      0.90       759\n",
      "        31.0       0.80      0.76      0.78       756\n",
      "        32.0       0.52      0.67      0.58       757\n",
      "        33.0       0.83      0.84      0.84       741\n",
      "        34.0       0.77      0.71      0.74       734\n",
      "        35.0       0.57      0.84      0.68       722\n",
      "        36.0       0.84      0.85      0.84       766\n",
      "        37.0       0.75      0.74      0.74       724\n",
      "        38.0       0.67      0.48      0.56       743\n",
      "        39.0       0.71      0.92      0.80       748\n",
      "        40.0       0.78      0.23      0.35       750\n",
      "        41.0       0.96      0.92      0.94       739\n",
      "        42.0       0.69      0.65      0.67       753\n",
      "        43.0       0.98      0.83      0.90       743\n",
      "        44.0       0.77      0.73      0.75       749\n",
      "        45.0       0.76      0.56      0.64       761\n",
      "        46.0       0.78      0.87      0.83       759\n",
      "        47.0       0.82      0.69      0.75       765\n",
      "        48.0       0.84      0.97      0.90       734\n",
      "        49.0       0.97      0.98      0.98       757\n",
      "        50.0       0.88      0.58      0.70       723\n",
      "        51.0       0.92      0.88      0.90       757\n",
      "        52.0       0.76      0.76      0.76       779\n",
      "        53.0       0.78      0.73      0.76       756\n",
      "        54.0       0.67      0.60      0.63       744\n",
      "        55.0       0.62      0.91      0.74       730\n",
      "        56.0       0.81      0.90      0.85       768\n",
      "        57.0       0.97      0.88      0.92       756\n",
      "        58.0       0.98      0.92      0.95       739\n",
      "        59.0       0.69      0.54      0.60       761\n",
      "        60.0       0.95      0.86      0.90       762\n",
      "        61.0       0.75      0.72      0.73       740\n",
      "        62.0       0.96      0.95      0.96       730\n",
      "        63.0       0.69      0.97      0.81       754\n",
      "        64.0       0.69      0.84      0.76       753\n",
      "        65.0       0.96      0.95      0.96       769\n",
      "        66.0       0.95      0.98      0.97       746\n",
      "        67.0       0.91      0.96      0.94       760\n",
      "        68.0       0.91      0.96      0.93       786\n",
      "        69.0       0.99      0.98      0.98       754\n",
      "        70.0       0.93      0.95      0.94       761\n",
      "        71.0       0.74      0.94      0.83       771\n",
      "        72.0       0.93      0.96      0.95       741\n",
      "        73.0       0.89      0.59      0.71       760\n",
      "        74.0       0.83      0.81      0.82       756\n",
      "        75.0       0.79      0.69      0.74       736\n",
      "        76.0       0.93      0.99      0.96       763\n",
      "        77.0       0.85      0.81      0.83       778\n",
      "        78.0       0.96      0.98      0.97       747\n",
      "        79.0       0.61      0.65      0.63       747\n",
      "        80.0       0.96      0.95      0.96       759\n",
      "        81.0       0.80      0.72      0.76       769\n",
      "        82.0       0.64      0.73      0.68       724\n",
      "        83.0       0.93      0.96      0.95       734\n",
      "        84.0       0.73      0.76      0.74       728\n",
      "        85.0       0.92      0.86      0.89       754\n",
      "        86.0       0.84      0.96      0.90       736\n",
      "        87.0       0.84      0.83      0.83       734\n",
      "        88.0       0.82      0.83      0.83       746\n",
      "        89.0       0.91      0.92      0.92       727\n",
      "        90.0       0.98      0.97      0.98       761\n",
      "        91.0       0.98      0.99      0.99       744\n",
      "        92.0       0.79      0.69      0.74       763\n",
      "        93.0       0.84      0.68      0.75       767\n",
      "        94.0       0.93      0.95      0.94       760\n",
      "        95.0       0.88      0.88      0.88       754\n",
      "        96.0       0.87      0.91      0.89       738\n",
      "        97.0       0.99      0.99      0.99       744\n",
      "        98.0       0.71      0.55      0.62       724\n",
      "        99.0       0.97      0.97      0.97       770\n",
      "       100.0       0.94      0.95      0.94       746\n",
      "\n",
      "    accuracy                           0.82     75000\n",
      "   macro avg       0.83      0.82      0.82     75000\n",
      "weighted avg       0.83      0.82      0.82     75000\n",
      "\n",
      "Classification Report:Testing\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         1.0       0.62      0.81      0.70       265\n",
      "         2.0       0.69      0.56      0.62       245\n",
      "         3.0       0.92      0.96      0.94       221\n",
      "         4.0       0.93      0.81      0.87       248\n",
      "         5.0       0.90      0.96      0.93       233\n",
      "         6.0       0.77      0.88      0.82       270\n",
      "         7.0       0.90      0.88      0.89       257\n",
      "         8.0       0.95      0.97      0.96       246\n",
      "         9.0       0.98      0.95      0.96       259\n",
      "        10.0       0.65      0.62      0.63       247\n",
      "        11.0       0.75      0.74      0.75       262\n",
      "        12.0       0.76      0.80      0.78       256\n",
      "        13.0       0.97      0.98      0.97       258\n",
      "        14.0       0.67      0.56      0.61       238\n",
      "        15.0       0.51      0.47      0.49       231\n",
      "        16.0       0.39      0.27      0.32       271\n",
      "        17.0       0.80      0.81      0.81       271\n",
      "        18.0       0.78      0.83      0.81       218\n",
      "        19.0       0.70      0.85      0.77       241\n",
      "        20.0       0.54      0.71      0.61       241\n",
      "        21.0       0.71      0.75      0.73       261\n",
      "        22.0       0.99      0.95      0.97       243\n",
      "        23.0       0.92      0.94      0.93       256\n",
      "        24.0       0.81      0.87      0.84       263\n",
      "        25.0       0.97      0.98      0.97       268\n",
      "        26.0       0.63      0.90      0.74       255\n",
      "        27.0       0.93      0.81      0.86       265\n",
      "        28.0       0.64      0.58      0.61       232\n",
      "        29.0       0.77      0.84      0.81       248\n",
      "        30.0       0.89      0.90      0.89       241\n",
      "        31.0       0.75      0.74      0.75       244\n",
      "        32.0       0.45      0.60      0.52       243\n",
      "        33.0       0.85      0.83      0.84       259\n",
      "        34.0       0.76      0.69      0.72       266\n",
      "        35.0       0.56      0.76      0.64       278\n",
      "        36.0       0.79      0.80      0.80       234\n",
      "        37.0       0.72      0.66      0.69       276\n",
      "        38.0       0.68      0.48      0.56       257\n",
      "        39.0       0.70      0.91      0.79       252\n",
      "        40.0       0.82      0.20      0.32       250\n",
      "        41.0       0.94      0.92      0.93       261\n",
      "        42.0       0.61      0.61      0.61       247\n",
      "        43.0       1.00      0.81      0.90       257\n",
      "        44.0       0.74      0.76      0.75       251\n",
      "        45.0       0.71      0.55      0.62       239\n",
      "        46.0       0.79      0.79      0.79       241\n",
      "        47.0       0.77      0.69      0.73       235\n",
      "        48.0       0.84      0.98      0.90       266\n",
      "        49.0       0.97      0.98      0.98       243\n",
      "        50.0       0.82      0.55      0.66       277\n",
      "        51.0       0.94      0.88      0.91       243\n",
      "        52.0       0.70      0.69      0.70       221\n",
      "        53.0       0.69      0.76      0.73       244\n",
      "        54.0       0.64      0.52      0.58       256\n",
      "        55.0       0.67      0.91      0.77       270\n",
      "        56.0       0.73      0.87      0.80       232\n",
      "        57.0       0.96      0.85      0.90       244\n",
      "        58.0       0.95      0.95      0.95       261\n",
      "        59.0       0.59      0.49      0.53       239\n",
      "        60.0       0.91      0.84      0.88       238\n",
      "        61.0       0.72      0.72      0.72       260\n",
      "        62.0       0.95      0.96      0.95       270\n",
      "        63.0       0.67      0.98      0.80       246\n",
      "        64.0       0.71      0.83      0.77       247\n",
      "        65.0       0.96      0.91      0.94       231\n",
      "        66.0       0.94      0.98      0.96       254\n",
      "        67.0       0.88      0.95      0.91       240\n",
      "        68.0       0.89      0.93      0.91       214\n",
      "        69.0       0.99      0.94      0.97       246\n",
      "        70.0       0.95      0.97      0.96       239\n",
      "        71.0       0.70      0.92      0.80       229\n",
      "        72.0       0.93      0.95      0.94       259\n",
      "        73.0       0.85      0.55      0.66       240\n",
      "        74.0       0.74      0.77      0.76       244\n",
      "        75.0       0.81      0.62      0.71       264\n",
      "        76.0       0.94      0.99      0.96       237\n",
      "        77.0       0.83      0.83      0.83       222\n",
      "        78.0       0.97      0.98      0.97       253\n",
      "        79.0       0.55      0.57      0.56       253\n",
      "        80.0       0.97      0.93      0.95       241\n",
      "        81.0       0.77      0.62      0.69       231\n",
      "        82.0       0.61      0.66      0.64       276\n",
      "        83.0       0.95      0.93      0.94       266\n",
      "        84.0       0.73      0.75      0.74       272\n",
      "        85.0       0.90      0.89      0.89       246\n",
      "        86.0       0.83      0.93      0.88       264\n",
      "        87.0       0.81      0.76      0.78       266\n",
      "        88.0       0.77      0.74      0.75       254\n",
      "        89.0       0.93      0.94      0.93       273\n",
      "        90.0       0.97      0.95      0.96       239\n",
      "        91.0       0.98      0.96      0.97       256\n",
      "        92.0       0.71      0.67      0.69       237\n",
      "        93.0       0.77      0.62      0.69       233\n",
      "        94.0       0.91      0.89      0.90       240\n",
      "        95.0       0.81      0.90      0.86       246\n",
      "        96.0       0.86      0.87      0.86       262\n",
      "        97.0       0.98      0.99      0.99       256\n",
      "        98.0       0.67      0.52      0.59       276\n",
      "        99.0       0.94      0.96      0.95       230\n",
      "       100.0       0.92      0.96      0.94       254\n",
      "\n",
      "    accuracy                           0.80     25000\n",
      "   macro avg       0.80      0.80      0.79     25000\n",
      "weighted avg       0.80      0.80      0.79     25000\n",
      "\n",
      "['MultinomialNB_model.pkl']\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\"\"\"  ------------ Best Model  -----------------\"\"\"\n",
    "\n",
    "from sklearn.naive_bayes import MultinomialNB \n",
    "\n",
    "model = MultinomialNB(alpha = 5)\n",
    "model.fit(X_train, Y_train)\n",
    "\n",
    "#Classification Report\n",
    "y_pred_train = model.predict(X_train)\n",
    "y_pred_test = model.predict(X_test)\n",
    "\n",
    "# print(\"\")\n",
    "print(\"Classification Report:Training\")\n",
    "print(classification_report(Y_train, y_pred_train))\n",
    "\n",
    "print(\"Classification Report:Testing\")\n",
    "print(classification_report(Y_test, y_pred_test))\n",
    "\n",
    "#saving the classifier \n",
    "import joblib\n",
    "\n",
    "joblib_file = \"MultinomialNB_model.pkl\"  \n",
    "joblib.dump(model, joblib_file)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bernoulli NB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "------------------ Results for Bernoulli NB ------------------\n",
      "\n",
      "alpha: 1e-08 testing: 0.7412   training: 0.80764\n",
      "alpha: 1e-07 testing: 0.74304   training: 0.8066133333333333\n",
      "alpha: 1e-06 testing: 0.74512   training: 0.8051866666666667\n",
      "alpha: 1e-05 testing: 0.74756   training: 0.80348\n",
      "alpha: 0.001 testing: 0.75228   training: 0.79828\n",
      "alpha: 0.01 testing: 0.75384   training: 0.7943066666666667\n",
      "alpha: 0.1 testing: 0.75232   training: 0.7876666666666666\n",
      "alpha: 0.5 testing: 0.74908   training: 0.7801466666666667\n",
      "alpha: 1 testing: 0.74688   training: 0.7748933333333333\n",
      "alpha: 5 testing: 0.73368   training: 0.7547733333333333\n",
      "alpha: 10 testing: 0.72096   training: 0.7403333333333333\n",
      "alpha: 50 testing: 0.62824   training: 0.64672\n",
      "alpha: 100 testing: 0.51908   training: 0.54408\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import BernoulliNB \n",
    "# arr stores the values of learning rates to be considered\n",
    "arr = [0.00000001,0.0000001, 0.000001,0.00001,0.001,0.01,0.1,0.5,1,5,10,50,100]\n",
    "\n",
    "\n",
    "train_score = []                                           # for storing train score for different values of learning rate\n",
    "test_score = []                                            # for storing test score for different values of learning rate  \n",
    "\n",
    "print(\"------------------ Results for Bernoulli NB ------------------\")\n",
    "\n",
    "for alpha in arr:\n",
    "    \n",
    "    # training the model on X_train \n",
    "    model = BernoulliNB(alpha = alpha, binarize = 0.5)\n",
    "    \n",
    "    model.fit(X_train, Y_train)\n",
    "    \n",
    "    # getting the accuracy on training and testing data\n",
    "    train_score_ = model.score(X_train, Y_train)\n",
    "    test_score_ = model.score(X_test, Y_test)\n",
    "    \n",
    "    # adding the train and test accuracy to train_score and test_score respectively\n",
    "    train_score.append(train_score_)\n",
    "    test_score.append(test_score_)\n",
    "\n",
    "    print(i,\"testing:\",test_score_,\" \",\"training:\",train_score_)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x19bb0a03ca0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "size = 10\n",
    "import matplotlib.pyplot as plt\n",
    "# plotting the loss vs iteration graph for training and validation data\n",
    "figure = plt.figure(1, figsize=(20,10))\n",
    "subplot  = figure.add_subplot(212)\n",
    "# subplot.tick_params(labelrotation=90)\n",
    "subplot.plot(arr[0:size], train_score[0:size], label = \"Training Accuracy\")\n",
    "subplot.plot(arr[0:size], test_score[0:size], label = \"Test Accuracy\")\n",
    "# subplot.scatter(X_iteration, Y_train_error)\n",
    "subplot.set_title(\"Accuracy vs Alpha, Model Used : BernoulliNB\")\n",
    "subplot.set_xlabel(\"Alpha\")\n",
    "subplot.set_ylabel(\"Accuracy\")\n",
    "subplot.legend(loc=\"upper right\")\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "76Dw3lXVvg75"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification Report:Training\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         1.0       0.64      0.75      0.69       735\n",
      "         2.0       0.72      0.57      0.64       755\n",
      "         3.0       0.95      0.93      0.94       779\n",
      "         4.0       0.89      0.83      0.86       752\n",
      "         5.0       0.95      0.93      0.94       767\n",
      "         6.0       0.70      0.89      0.78       730\n",
      "         7.0       0.93      0.87      0.90       743\n",
      "         8.0       0.99      0.97      0.98       754\n",
      "         9.0       0.98      0.93      0.95       741\n",
      "        10.0       0.63      0.60      0.62       753\n",
      "        11.0       0.73      0.80      0.76       738\n",
      "        12.0       0.76      0.68      0.72       744\n",
      "        13.0       0.98      0.97      0.97       742\n",
      "        14.0       0.84      0.56      0.68       762\n",
      "        15.0       0.58      0.57      0.58       769\n",
      "        16.0       0.27      0.32      0.29       729\n",
      "        17.0       0.81      0.82      0.81       729\n",
      "        18.0       0.85      0.81      0.83       782\n",
      "        19.0       0.71      0.85      0.78       759\n",
      "        20.0       0.58      0.77      0.66       759\n",
      "        21.0       0.72      0.78      0.75       739\n",
      "        22.0       1.00      0.92      0.96       757\n",
      "        23.0       0.93      0.95      0.94       744\n",
      "        24.0       0.74      0.85      0.79       737\n",
      "        25.0       0.96      0.93      0.95       732\n",
      "        26.0       0.65      0.91      0.76       745\n",
      "        27.0       0.89      0.86      0.87       735\n",
      "        28.0       0.43      0.74      0.54       768\n",
      "        29.0       0.87      0.84      0.86       752\n",
      "        30.0       0.91      0.85      0.88       759\n",
      "        31.0       0.71      0.72      0.72       756\n",
      "        32.0       0.46      0.66      0.54       757\n",
      "        33.0       0.35      0.92      0.51       741\n",
      "        34.0       0.82      0.68      0.74       734\n",
      "        35.0       0.62      0.73      0.67       722\n",
      "        36.0       0.85      0.82      0.84       766\n",
      "        37.0       0.76      0.60      0.67       724\n",
      "        38.0       0.56      0.49      0.53       743\n",
      "        39.0       0.72      0.88      0.79       748\n",
      "        40.0       0.99      0.37      0.54       750\n",
      "        41.0       0.95      0.89      0.92       739\n",
      "        42.0       0.73      0.58      0.65       753\n",
      "        43.0       0.99      0.78      0.87       743\n",
      "        44.0       0.76      0.77      0.76       749\n",
      "        45.0       0.76      0.54      0.63       761\n",
      "        46.0       0.82      0.88      0.85       759\n",
      "        47.0       0.74      0.68      0.71       765\n",
      "        48.0       0.82      0.93      0.87       734\n",
      "        49.0       0.99      0.94      0.96       757\n",
      "        50.0       0.90      0.46      0.61       723\n",
      "        51.0       0.92      0.87      0.89       757\n",
      "        52.0       0.74      0.63      0.68       779\n",
      "        53.0       0.79      0.70      0.74       756\n",
      "        54.0       0.75      0.59      0.66       744\n",
      "        55.0       0.68      0.88      0.77       730\n",
      "        56.0       0.67      0.91      0.78       768\n",
      "        57.0       0.97      0.88      0.92       756\n",
      "        58.0       0.94      0.94      0.94       739\n",
      "        59.0       0.63      0.48      0.54       761\n",
      "        60.0       0.92      0.76      0.84       762\n",
      "        61.0       0.78      0.71      0.74       740\n",
      "        62.0       0.96      0.89      0.93       730\n",
      "        63.0       0.74      0.94      0.83       754\n",
      "        64.0       0.69      0.86      0.77       753\n",
      "        65.0       1.00      0.88      0.93       769\n",
      "        66.0       0.92      0.98      0.95       746\n",
      "        67.0       0.94      0.92      0.93       760\n",
      "        68.0       0.89      0.96      0.92       786\n",
      "        69.0       1.00      0.93      0.96       754\n",
      "        70.0       0.94      0.93      0.93       761\n",
      "        71.0       0.76      0.92      0.83       771\n",
      "        72.0       0.92      0.90      0.91       741\n",
      "        73.0       0.89      0.51      0.65       760\n",
      "        74.0       0.81      0.78      0.80       756\n",
      "        75.0       0.85      0.65      0.74       736\n",
      "        76.0       0.96      0.96      0.96       763\n",
      "        77.0       0.86      0.77      0.81       778\n",
      "        78.0       0.97      0.95      0.96       747\n",
      "        79.0       0.53      0.56      0.54       747\n",
      "        80.0       0.98      0.90      0.93       759\n",
      "        81.0       0.75      0.58      0.65       769\n",
      "        82.0       0.66      0.62      0.64       724\n",
      "        83.0       0.93      0.92      0.93       734\n",
      "        84.0       0.77      0.77      0.77       728\n",
      "        85.0       0.93      0.76      0.84       754\n",
      "        86.0       0.81      0.91      0.86       736\n",
      "        87.0       0.85      0.82      0.83       734\n",
      "        88.0       0.83      0.80      0.81       746\n",
      "        89.0       0.82      0.91      0.86       727\n",
      "        90.0       0.99      0.96      0.97       761\n",
      "        91.0       1.00      0.94      0.97       744\n",
      "        92.0       0.72      0.72      0.72       763\n",
      "        93.0       0.83      0.75      0.79       767\n",
      "        94.0       0.83      0.99      0.90       760\n",
      "        95.0       0.89      0.82      0.85       754\n",
      "        96.0       0.87      0.86      0.87       738\n",
      "        97.0       1.00      0.97      0.99       744\n",
      "        98.0       0.76      0.55      0.64       724\n",
      "        99.0       0.91      0.97      0.94       770\n",
      "       100.0       0.97      0.90      0.93       746\n",
      "\n",
      "    accuracy                           0.79     75000\n",
      "   macro avg       0.81      0.79      0.80     75000\n",
      "weighted avg       0.81      0.79      0.80     75000\n",
      "\n",
      "Classification Report:Testing\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "         1.0       0.67      0.80      0.73       265\n",
      "         2.0       0.64      0.54      0.59       245\n",
      "         3.0       0.94      0.92      0.93       221\n",
      "         4.0       0.92      0.79      0.85       248\n",
      "         5.0       0.91      0.94      0.92       233\n",
      "         6.0       0.66      0.83      0.74       270\n",
      "         7.0       0.91      0.81      0.86       257\n",
      "         8.0       0.98      0.96      0.97       246\n",
      "         9.0       0.97      0.88      0.92       259\n",
      "        10.0       0.54      0.54      0.54       247\n",
      "        11.0       0.66      0.70      0.68       262\n",
      "        12.0       0.74      0.64      0.68       256\n",
      "        13.0       0.98      0.97      0.97       258\n",
      "        14.0       0.74      0.59      0.66       238\n",
      "        15.0       0.47      0.47      0.47       231\n",
      "        16.0       0.23      0.32      0.26       271\n",
      "        17.0       0.79      0.75      0.77       271\n",
      "        18.0       0.79      0.81      0.80       218\n",
      "        19.0       0.68      0.84      0.75       241\n",
      "        20.0       0.50      0.70      0.58       241\n",
      "        21.0       0.71      0.72      0.71       261\n",
      "        22.0       0.99      0.90      0.94       243\n",
      "        23.0       0.95      0.95      0.95       256\n",
      "        24.0       0.70      0.86      0.77       263\n",
      "        25.0       0.96      0.94      0.95       268\n",
      "        26.0       0.63      0.86      0.73       255\n",
      "        27.0       0.95      0.79      0.86       265\n",
      "        28.0       0.35      0.66      0.46       232\n",
      "        29.0       0.83      0.74      0.78       248\n",
      "        30.0       0.90      0.82      0.86       241\n",
      "        31.0       0.60      0.65      0.62       244\n",
      "        32.0       0.41      0.60      0.49       243\n",
      "        33.0       0.35      0.88      0.50       259\n",
      "        34.0       0.77      0.64      0.70       266\n",
      "        35.0       0.59      0.64      0.61       278\n",
      "        36.0       0.78      0.74      0.76       234\n",
      "        37.0       0.71      0.56      0.63       276\n",
      "        38.0       0.55      0.42      0.47       257\n",
      "        39.0       0.73      0.84      0.78       252\n",
      "        40.0       0.98      0.37      0.53       250\n",
      "        41.0       0.96      0.87      0.91       261\n",
      "        42.0       0.58      0.45      0.51       247\n",
      "        43.0       1.00      0.74      0.85       257\n",
      "        44.0       0.70      0.74      0.72       251\n",
      "        45.0       0.63      0.50      0.56       239\n",
      "        46.0       0.82      0.76      0.79       241\n",
      "        47.0       0.69      0.64      0.67       235\n",
      "        48.0       0.79      0.92      0.85       266\n",
      "        49.0       1.00      0.92      0.96       243\n",
      "        50.0       0.87      0.40      0.55       277\n",
      "        51.0       0.94      0.81      0.87       243\n",
      "        52.0       0.66      0.57      0.62       221\n",
      "        53.0       0.71      0.71      0.71       244\n",
      "        54.0       0.63      0.46      0.53       256\n",
      "        55.0       0.72      0.86      0.78       270\n",
      "        56.0       0.56      0.88      0.69       232\n",
      "        57.0       0.97      0.85      0.91       244\n",
      "        58.0       0.92      0.95      0.93       261\n",
      "        59.0       0.51      0.44      0.47       239\n",
      "        60.0       0.83      0.74      0.78       238\n",
      "        61.0       0.73      0.70      0.71       260\n",
      "        62.0       0.96      0.87      0.91       270\n",
      "        63.0       0.73      0.92      0.81       246\n",
      "        64.0       0.71      0.83      0.77       247\n",
      "        65.0       0.97      0.85      0.91       231\n",
      "        66.0       0.94      0.97      0.96       254\n",
      "        67.0       0.88      0.89      0.89       240\n",
      "        68.0       0.85      0.94      0.89       214\n",
      "        69.0       1.00      0.90      0.95       246\n",
      "        70.0       0.98      0.94      0.96       239\n",
      "        71.0       0.70      0.90      0.79       229\n",
      "        72.0       0.91      0.89      0.90       259\n",
      "        73.0       0.81      0.43      0.56       240\n",
      "        74.0       0.71      0.68      0.69       244\n",
      "        75.0       0.74      0.54      0.63       264\n",
      "        76.0       0.97      0.94      0.95       237\n",
      "        77.0       0.82      0.78      0.80       222\n",
      "        78.0       0.98      0.95      0.96       253\n",
      "        79.0       0.44      0.50      0.47       253\n",
      "        80.0       0.97      0.88      0.92       241\n",
      "        81.0       0.56      0.45      0.50       231\n",
      "        82.0       0.59      0.51      0.54       276\n",
      "        83.0       0.93      0.91      0.92       266\n",
      "        84.0       0.75      0.72      0.74       272\n",
      "        85.0       0.92      0.76      0.83       246\n",
      "        86.0       0.80      0.92      0.85       264\n",
      "        87.0       0.80      0.74      0.77       266\n",
      "        88.0       0.69      0.70      0.70       254\n",
      "        89.0       0.83      0.91      0.87       273\n",
      "        90.0       0.97      0.92      0.94       239\n",
      "        91.0       0.98      0.91      0.95       256\n",
      "        92.0       0.66      0.69      0.67       237\n",
      "        93.0       0.77      0.64      0.69       233\n",
      "        94.0       0.79      0.92      0.85       240\n",
      "        95.0       0.84      0.84      0.84       246\n",
      "        96.0       0.85      0.79      0.82       262\n",
      "        97.0       1.00      0.94      0.97       256\n",
      "        98.0       0.68      0.49      0.57       276\n",
      "        99.0       0.83      0.91      0.87       230\n",
      "       100.0       0.96      0.91      0.93       254\n",
      "\n",
      "    accuracy                           0.75     25000\n",
      "   macro avg       0.78      0.75      0.76     25000\n",
      "weighted avg       0.78      0.75      0.76     25000\n",
      "\n",
      "['BernoulliNB_model.pkl']\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\"\"\"  ------------ Best Model  -----------------\"\"\"\n",
    "\n",
    "from sklearn.naive_bayes import BernoulliNB \n",
    "\n",
    "model = BernoulliNB(alpha = 0.01, binarize = 0.5)\n",
    "model.fit(X_train, Y_train)\n",
    "\n",
    "#Classification Report\n",
    "y_pred_train = model.predict(X_train)\n",
    "y_pred_test = model.predict(X_test)\n",
    "\n",
    "# print(\"\")\n",
    "print(\"Classification Report:Training\")\n",
    "print(classification_report(Y_train, y_pred_train))\n",
    "\n",
    "print(\"Classification Report:Testing\")\n",
    "print(classification_report(Y_test, y_pred_test))\n",
    "\n",
    "#saving the classifier \n",
    "import joblib\n",
    "\n",
    "joblib_file = \"BernoulliNB_model.pkl\"  \n",
    "joblib.dump(model, joblib_file)\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Tony007.ipynb",
   "provenance": []
  },
  "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}