naive_baesian.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalize(X):\n",
    "    for i in range(X.shape[1]):\n",
    "        X[:,i] = (X[:,i] - X[:,i].mean())/X[:,i].std()\n",
    "    return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      1     2     3     4    5    6     7     8 OutName\n",
      "0  0.58  0.61  0.47  0.13  0.5  0.0  0.48  0.22     MIT\n",
      "1  0.43  0.67  0.48  0.27  0.5  0.0  0.53  0.22     MIT\n",
      "2  0.64  0.62  0.49  0.15  0.5  0.0  0.53  0.22     MIT\n",
      "3  0.58  0.44  0.57  0.13  0.5  0.0  0.54  0.22     NUC\n",
      "4  0.42  0.44  0.48  0.54  0.5  0.0  0.48  0.22     MIT\n",
      "5  0.51  0.40  0.56  0.17  0.5  0.5  0.49  0.22     CYT\n",
      "6  0.50  0.54  0.48  0.65  0.5  0.0  0.53  0.22     MIT\n",
      "7  0.48  0.45  0.59  0.20  0.5  0.0  0.58  0.34     NUC\n",
      "8  0.55  0.50  0.66  0.36  0.5  0.0  0.49  0.22     MIT\n",
      "9  0.40  0.39  0.60  0.15  0.5  0.0  0.58  0.30     CYT\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# from sklearn import tree\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "def get_data():\n",
    "    \"\"\"\n",
    "    Returns the data from the xlsx file\n",
    "    \"\"\"\n",
    "    file_name = 'yeast.csv' \n",
    "    df = pd.read_csv(file_name, index_col=None)\n",
    "    # print(df.head()) # print the first 5 rows\n",
    "    return df\n",
    "\n",
    "C = 10\n",
    "\n",
    "D = 8\n",
    "# p = 8\n",
    "\n",
    "md = get_data()\n",
    "\n",
    "\n",
    "# drop the 0 column\n",
    "# ?????????\n",
    "md = md.drop(columns=['DBName'])\n",
    "\n",
    "md.dropna(inplace = True)\n",
    "# print(md.shape)\n",
    "md.replace('?', 0, inplace = True)\n",
    "\n",
    "print(md.head(10))\n",
    "\n",
    "# print(md[0 : 1])\n",
    "\n",
    "#print(columns[1])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert continuous data to categorical one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>cat1</th>\n",
       "      <th>cat2</th>\n",
       "      <th>cat3</th>\n",
       "      <th>cat4</th>\n",
       "      <th>cat5</th>\n",
       "      <th>cat6</th>\n",
       "      <th>cat7</th>\n",
       "      <th>cat8</th>\n",
       "      <th>OutName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>26</td>\n",
       "      <td>27</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17</td>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>29</td>\n",
       "      <td>28</td>\n",
       "      <td>17</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>26</td>\n",
       "      <td>17</td>\n",
       "      <td>22</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>10</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>22</td>\n",
       "      <td>15</td>\n",
       "      <td>22</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>33</td>\n",
       "      <td>10</td>\n",
       "      <td>CYT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>21</td>\n",
       "      <td>23</td>\n",
       "      <td>17</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>20</td>\n",
       "      <td>18</td>\n",
       "      <td>24</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "      <td>16</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>24</td>\n",
       "      <td>21</td>\n",
       "      <td>28</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>16</td>\n",
       "      <td>14</td>\n",
       "      <td>24</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "      <td>14</td>\n",
       "      <td>CYT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>17</td>\n",
       "      <td>14</td>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>13</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>17</td>\n",
       "      <td>13</td>\n",
       "      <td>24</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>14</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>22</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>12</td>\n",
       "      <td>CYT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>27</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>30</td>\n",
       "      <td>24</td>\n",
       "      <td>15</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>10</td>\n",
       "      <td>MIT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>19</td>\n",
       "      <td>17</td>\n",
       "      <td>19</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>10</td>\n",
       "      <td>CYT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>20</td>\n",
       "      <td>14</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>19</td>\n",
       "      <td>CYT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>26</td>\n",
       "      <td>19</td>\n",
       "      <td>20</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>12</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>21</td>\n",
       "      <td>12</td>\n",
       "      <td>21</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "      <td>10</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>28</td>\n",
       "      <td>27</td>\n",
       "      <td>21</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>12</td>\n",
       "      <td>NUC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   cat1 cat2 cat3 cat4 cat5 cat6 cat7 cat8 OutName\n",
       "0    26   27   16    6    0    0   32   10     MIT\n",
       "1    17   31   17   13    0    0   36   10     MIT\n",
       "2    29   28   17    7    0    0   36   10     MIT\n",
       "3    26   17   22    6    0    0   36   10     NUC\n",
       "4    17   17   17   26    0    0   32   10     MIT\n",
       "5    22   15   22    8    0   30   33   10     CYT\n",
       "6    21   23   17   32    0    0   36   10     MIT\n",
       "7    20   18   24    9    0    0   39   16     NUC\n",
       "8    24   21   28   17    0    0   33   10     MIT\n",
       "9    16   14   24    7    0    0   39   14     CYT\n",
       "10   17   14   20   10    0    0   36   13     NUC\n",
       "11   17   13   24    9    0    0   35   14     NUC\n",
       "12   16   16   22   17    0    0   36   12     CYT\n",
       "13   27   15   19   22    0    0   36   10     MIT\n",
       "14   30   24   15    9    0    0   31   10     MIT\n",
       "15   19   17   19    5    0    0   34   10     CYT\n",
       "16   20   14   18    5    0    0   33   19     CYT\n",
       "17   26   19   20    5    0    0   34   12     NUC\n",
       "18   21   12   21   10    0    0   33   10     NUC\n",
       "19   28   27   21   10    0    0   34   12     NUC"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "noOfBins = 50\n",
    "# print(md['DBName'])\n",
    "# print(md.columns)\n",
    "# print(md['1'])\n",
    "for i in range(D):\n",
    "    md[\"cat\" + str(i + 1)] = pd.cut(md[str(i+1)].values, bins=noOfBins, labels= range(0, noOfBins))\n",
    "md.drop(columns=['1', '2', '3', '4', '5', '6', '7', '8'], inplace=True)\n",
    "last_column = md.pop('OutName')\n",
    "md.insert(8, 'OutName', last_column)\n",
    "md.head(20)\n",
    "\n",
    "\n",
    "# gk = md.groupby('Outname')\n",
    "# print(md[1].value_counts())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# X = md.iloc[:, 0:8].values.reshape(-1, 8)\n",
    "\n",
    "outputs = ['CYT', 'NUC', 'MIT', 'ME3', 'ME2', 'ME1', 'EXC', 'VAC', 'POX', 'ERL']\n",
    "\n",
    "# outputs = ['CYT', 'ERL', 'EXC', 'ME1', 'ME2', 'ME3', 'MIT', 'NUC', 'POX', 'VAC']\n",
    "\n",
    "# M = X.shape[0]\n",
    "# Y = md.iloc[:, 8:9].values.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# y = np.zeros(len(Y))\n",
    "\n",
    "# for i in range(len(Y)):\n",
    "#     y[i] = outputs.index(Y[i][0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "trainSize = 1200\n",
    "\n",
    "def calculatePiAndTheta(md, C, alpha):\n",
    "    ncValues = np.zeros(C)\n",
    "    piValues = []\n",
    "    \n",
    "    Y = md.iloc[:, 8:9].values.reshape(-1, 1)\n",
    "\n",
    "    y = np.zeros(len(Y))\n",
    "\n",
    "    for i in range(len(Y)):\n",
    "        y[i] = outputs.index(Y[i][0])\n",
    "\n",
    "    testSize = len(md)\n",
    "\n",
    "    # calculate the pi values\n",
    "    for i in range(testSize):\n",
    "        ncValues[int(y[i])] += 1\n",
    "    for c in range(C):\n",
    "        piValues.append(ncValues[c]/testSize)\n",
    "\n",
    "    # calculate theta values\n",
    "    md[['cat1', 'cat2', 'cat3', 'cat4', 'cat5', 'cat6', 'cat7', 'cat8']] = md[['cat1', 'cat2', 'cat3', 'cat4', 'cat5', 'cat6', 'cat7', 'cat8']].astype(int)\n",
    "\n",
    "    data_mod = md.groupby(['OutName']).sum()\n",
    "\n",
    "    # arrange colums w.r.t. hardcoded output array\n",
    "    data_mod.unstack()\n",
    "    data_mod_t = data_mod.transpose()\n",
    "    data_mod_t = data_mod_t[outputs]\n",
    "    data_mod = data_mod_t.transpose()\n",
    "\n",
    "    axisSum = data_mod.sum(axis=1)\n",
    "\n",
    "    theta = []\n",
    "\n",
    "    for d in range(D):\n",
    "        theta.append((data_mod['cat' + str(d + 1)] + alpha )/ (axisSum + alpha*D))\n",
    "\n",
    "    return piValues, theta\n",
    "alpha = 0.01\n",
    "\n",
    "\n",
    "train = md.copy()\n",
    "\n",
    "train = train.iloc[:trainSize, :]\n",
    "\n",
    "\n",
    "\n",
    "piValues, theta = calculatePiAndTheta(train, C, alpha)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:  0.528169014084507\n",
      "150 284\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def predict(md, piValues, theta, C, D):\n",
    "    # testSet = X\n",
    "    X = md.iloc[:, 0:8].values.reshape(-1, 8)\n",
    "    Y = md.iloc[:, 8:9].values.reshape(-1, 1)\n",
    "\n",
    "    y = np.zeros(len(Y))\n",
    "\n",
    "    for i in range(len(Y)):\n",
    "        y[i] = outputs.index(Y[i][0])\n",
    "\n",
    "\n",
    "    prediction = np.zeros(len(X))\n",
    "\n",
    "    for i in range(len(X)): \n",
    "        prbAllClass = np.zeros(C)\n",
    "        for c in range(C):\n",
    "            prbAllClass[c] = np.log(piValues[c])\n",
    "            for d in range(D):\n",
    "                first = X[i][d]\n",
    "                \n",
    "                # first = np.log(1 + md_dic['cat' + str(d + 1)][i])\n",
    "                second = np.log(theta[d][c]) # ['cat' + str(d + 1)][outputs[c]])\n",
    "                \n",
    "                prbAllClass[c] += first * second\n",
    "\n",
    "        prediction[i] = np.argmax(prbAllClass)\n",
    "\n",
    "    # print(\"accuracy: \", (prediction == y[:len(X)]).sum()/len(X))\n",
    "    return prediction, (prediction == y[:len(X)]).sum(), len(X)\n",
    "\n",
    "\n",
    "test = md.copy()\n",
    "test = test.iloc[trainSize:, :]\n",
    "\n",
    "result = predict(test, piValues, theta, C, D)\n",
    "print(\"accuracy: \", result[1] / result[2])\n",
    "print(result[1], result[2])\n",
    "\n",
    "\n",
    "# print confusion matrix\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "import matplotlib.pyplot as plt\n",
    "y_pred_list = []\n",
    "y_pred = result[0]\n",
    "Y_test = test.iloc[:, 8:9].values.reshape(-1, 1)\n",
    "y_test = np.zeros(len(Y_test))\n",
    "for i in range(len(Y_test)):\n",
    "    y_test[i] = outputs.index(Y_test[i][0])\n",
    "\n",
    "\n",
    "for i in range(len(test)):\n",
    "    y_pred_list.append(outputs[int(y_pred[i])])\n",
    "ConfusionMatrixDisplay.from_predictions(Y_test, y_pred_list)\n",
    "plt.show()\n",
    "\n",
    "# from sklearn.metrics import classification_report\n",
    "# classification_report(Y_test, y_pred_list, zero_division=1)\n",
    "# print('              precision    recall  f1-score   support\\n\\n         CYT       0.61      0.58      0.59        99\\n         EXC       0.31      1.00      0.48         5\\n         ME1       0.65      0.65      0.65        20\\n         ME2       0.29      0.25      0.27         8\\n         ME3       0.81      0.51      0.62        49\\n         MIT       0.24      0.38      0.30        21\\n         NUC       0.47      0.60      0.53        63\\n         POX       1.00      0.50      0.67         4\\n         VAC       0.00      0.00      0.00        15\\n\\n    accuracy                           0.53       284\\n   macro avg       0.49      0.50      0.46       284\\nweighted avg       0.55      0.53      0.53       284\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.01\n",
      "accuracy:  0.5633333333333334\n"
     ]
    }
   ],
   "source": [
    "# test for different alpha values \n",
    "# apply cross one out validation to compute accuracy\n",
    "\n",
    "# alphaValues = [0.001, 0.01, 0.05, 0.1, 1, 2, 5, 10, 100]\n",
    "alphaValues = [0.01]\n",
    "\n",
    "for alpha in alphaValues:\n",
    "    print(alpha)\n",
    "    correct = 0\n",
    "    for i in range(len(train)):\n",
    "        data = train.drop(axis=0, index=i)\n",
    "\n",
    "        cutted = train.iloc[i]\n",
    "        cuttedFrame = cutted.to_frame()\n",
    "        cuttedFrame = cuttedFrame.transpose()\n",
    "\n",
    "        y_test = outputs.index(train.iloc[i]['OutName'])\n",
    "        y_test = np.array( [y_test] )\n",
    "\n",
    "        x_test = np.zeros((1, D))\n",
    "        for d in range(D):\n",
    "            x_test[0][d] = cutted['cat' + str(d + 1)]\n",
    "\n",
    "        X = data.iloc[:, 0:8].values.reshape(-1, 8)\n",
    "        Y = data.iloc[:, 8:9].values.reshape(-1, 1)\n",
    "\n",
    "        y = np.zeros(len(Y))\n",
    "        for i in range(len(Y)):\n",
    "            y[i] = outputs.index(Y[i][0])\n",
    "\n",
    "        piValues, theta = calculatePiAndTheta(data, C, alpha)\n",
    "        res = predict(cuttedFrame, piValues, theta, C, D)\n",
    "        \n",
    "\n",
    "        correct += res[1]\n",
    "\n",
    "    print(\"accuracy: \", correct/len(train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from sklearn.model_selection import train_test_split\n",
    "# from sklearn.naive_bayes import ComplementNB\n",
    "\n",
    "# X_train, X_test, Y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)\n",
    "# gnb = ComplementNB()\n",
    "# gnb.fit(X_train, Y_train)\n",
    "# y_pred = gnb.predict(X_test)\n",
    "# print(\"Number of mislabeled points out of a total %d points : %d\"\n",
    "#       % (X_test.shape[0], (y_test != y_pred).sum()))\n",
    "\n",
    "# from sklearn import metrics\n",
    "\n",
    "# # Model Accuracy, how often is the classifier correct?\n",
    "# print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of mislabeled points out of a total 480 points : 219\n",
      "Accuracy: 0.54375\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, Y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)\n",
    "\n",
    "gnb = MultinomialNB()\n",
    "gnb.fit(X_train, Y_train)\n",
    "\n",
    "# X_test = X[:50]\n",
    "y_pred = gnb.predict(X_test)\n",
    "# y_test = y[:50]\n",
    "\n",
    "\n",
    "print(\"Number of mislabeled points out of a total %d points : %d\"\n",
    "      % (X_test.shape[0], (y_test != y_pred).sum()))\n",
    "\n",
    "\n",
    "\n",
    "from sklearn import metrics\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\",metrics.accuracy_score(y_test, y_pred))\n"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "03aca06cf7a7f4ecfa091c2aa2bbd40553b40c2a92bb9cf1b5b673de927a26ff"
  },
  "kernelspec": {
   "display_name": "Python 3.10.4 64-bit (windows store)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.4"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}