regogniser.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.cluster import k_means\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.svm import SVC\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the test_csv and train_csv\n",
    "test_csv = pd.read_csv(\"test.csv\")\n",
    "train_csv = pd.read_csv(\"train.csv\")\n",
    "sample_csv = pd.read_csv(\"sample_submission.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(784,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# displaying the number in pixelated form as image (vector points to image for our convenience)\n",
    "\n",
    "df = pd.concat((train_csv, test_csv))\n",
    "\n",
    "features = df.columns.drop('label')\n",
    "print(features.shape)\n",
    "\n",
    "\n",
    "dum = pd.get_dummies(df, columns=['label'])\n",
    "df = pd.concat([df['label'], dum]) \n",
    "\n",
    "arr = np.asarray(np.array(train_csv.loc[np.random.randint(0, 784)][1:]).reshape(-28, 28))\n",
    "plt.imshow(arr, cmap='inferno', vmin=0, vmax=255)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = train_csv.drop(['label'], axis=1)\n",
    "y_train = train_csv['label']\n",
    "\n",
    "X_train = X_train.values.reshape(-784,784)\n",
    "test_csv = (test_csv).values.reshape(-28,28)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = y_train.values.reshape(-1,1)\n",
    "y_train = OneHotEncoder(sparse=False).fit_transform(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train test split of X_train and y_train with reshaping\n",
    "KX_train, KX_test, KY_train, KY_test = train_test_split(X_train, y_train, test_size=0.2, random_state=452)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x22b2b7eaf20>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Image representation of the data after preprocessing\n",
    "plt.imshow(KX_train[np.random.randint(100)].reshape(-28, 28), cmap='inferno', vmin=0, vmax=255)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9930238095238095\n"
     ]
    }
   ],
   "source": [
    "# training the madel using KNN classifier\n",
    "knn = KNeighborsClassifier(n_neighbors=6)\n",
    "knn.fit(KX_train, KY_train)\n",
    "\n",
    "# predicting the test data\n",
    "pred = knn.predict(KX_test)\n",
    "\n",
    "# accuracy of the model\n",
    "print(np.mean(pred == KY_test))\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.3 64-bit",
   "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": "3.10.3"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "37d127deac291ffcf9c7f40aed0f25c4c2b1eb59582edad5fe8fcfda13ce033a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}