digit_recognition(random).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import the libraries and input the data\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "train_df = pd.read_csv(\"train.csv\")\n",
    "test_df = pd.read_csv(\"test.csv\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "##extracting Independent and dependent variable\n",
    "X = train_df.drop('label', axis= 1)\n",
    "Y = train_df['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model training\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train , X_test , Y_train, Y_test = train_test_split(X, Y, test_size= 0.3, random_state=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data preprocessing\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "sc= StandardScaler()\n",
    "\n",
    "#fit transform use to do some calculation and do the transformation\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC()"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##model construction\n",
    "from sklearn.svm import SVC\n",
    "rbf_model = SVC(kernel='rbf')\n",
    "rbf_model.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#model testing\n",
    "y_rbf_pred = rbf_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted values :\n",
      " [3 9 6 7 1]\n",
      "Actual Values :\n",
      " 24273    3\n",
      "32691    9\n",
      "34526    6\n",
      "11625    7\n",
      "6614     1\n",
      "Name: label, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"Predicted values :\\n\", y_rbf_pred[10:15])\n",
    "print(\"Actual Values :\\n\", Y_test[10:15])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:  0.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.9565870.956587 %\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "acc_rbf = metrics.accuracy_score(Y_test , y_rbf_pred)\n",
    "print(\"accuracy: \", \"{:2f}\".format(acc_rbf)*100, \"%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#using random integers for testing\n",
    "for i in (np.random.randint(0,270,4)):\n",
    "    two_d = (np.reshape(X_test[i], (28,28)))\n",
    "    plt.title('predicted label :{0}'.format(y_rbf_pred[i]))\n",
    "    plt.imshow(two_d, cmap='gray')\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}