naive-bayes.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Naive Bayes.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyN50IGHqWpzzXaChRs/msOy",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/osamazafar980/6a32aaf1d3d148e75da101124e25f1bd/naive-bayes.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UNPZgcTdspYK",
        "colab_type": "text"
      },
      "source": [
        "Importing Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ytZaNWWFsu7R",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from sklearn.datasets import fetch_openml\n",
        "%matplotlib inline"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XcyUh-Ssph9A",
        "colab_type": "text"
      },
      "source": [
        "# Preparing Dataset\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lEv31aNuptkk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 498
        },
        "outputId": "8b19ff23-18ed-42c9-90f8-43cf72743848"
      },
      "source": [
        "mnist = fetch_openml('mnist_784', version=1)\n",
        "x = mnist['data']\n",
        "y = mnist['target']\n",
        "print(x.dtype, y.dtype)\n",
        "print(x.shape,y.shape)\n",
        "\n",
        "#Ploting Data\n",
        "\n",
        "def plot_images(images , labels):\n",
        "  n_cols = min(5,len(images))\n",
        "  n_rows = len(images)//n_cols\n",
        "  fig = plt.figure(figsize=(8,8))\n",
        "\n",
        "  for i  in range (n_rows * n_cols):\n",
        "    sp = fig.add_subplot(n_rows,n_cols,i+1)\n",
        "    plt.axis(\"off\")\n",
        "    plt.imshow(images[i], cmap=plt.cm.gray)\n",
        "    sp.set_title(labels[i])\n",
        "  plt.show()\n",
        "\n",
        "p = np.random.permutation(len(x))\n",
        "p = p[:20]\n",
        "plot_images(x[p].reshape(-1,28,28),y[p])\n",
        "\n"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "float64 object\n",
            "(70000, 784) (70000,)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcwAAAG/CAYAAADVbefpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nD5pJsLpLRo0xAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 576x576 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OGJLrtMf2eVU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "8da18033-ae1e-4ef5-b8fb-a2398c8c928e"
      },
      "source": [
        "y=y.astype(\"int32\")\n",
        "x = x/255.0\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "tx,tesx,ty,tesy = train_test_split(x,y)\n",
        "tx.shape,tesx.shape"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((52500, 784), (17500, 784))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F_C-dC9r3k4U",
        "colab_type": "text"
      },
      "source": [
        "Training Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FkkF0peX3nj9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "76e525b7-0335-4a66-f1ff-8507411c4efd"
      },
      "source": [
        "from sklearn.naive_bayes import BernoulliNB\n",
        "\n",
        "cls = BernoulliNB()\n",
        "cls.fit(tx,ty)\n"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BernoulliNB(alpha=1.0, binarize=0.0, class_prior=None, fit_prior=True)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_MGcAZhH341F",
        "colab_type": "text"
      },
      "source": [
        "Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e02XPp7P365d",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 319
        },
        "outputId": "c7f3ab56-6024-448f-b17f-17dc4fb661d4"
      },
      "source": [
        "print(cls.score(tesx,tesy))\n",
        "from sklearn.metrics import classification_report\n",
        "pre = cls.predict(tesx)\n",
        "print(classification_report(tesy,pre))\n"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.8305714285714285\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.92      0.89      0.91      1784\n",
            "           1       0.87      0.95      0.91      1920\n",
            "           2       0.86      0.82      0.84      1753\n",
            "           3       0.75      0.80      0.78      1817\n",
            "           4       0.82      0.80      0.81      1640\n",
            "           5       0.79      0.69      0.74      1550\n",
            "           6       0.90      0.90      0.90      1720\n",
            "           7       0.91      0.84      0.88      1840\n",
            "           8       0.77      0.76      0.77      1723\n",
            "           9       0.73      0.82      0.77      1753\n",
            "\n",
            "    accuracy                           0.83     17500\n",
            "   macro avg       0.83      0.83      0.83     17500\n",
            "weighted avg       0.83      0.83      0.83     17500\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}