K_Nearest_Neighbors_(knn).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "598aa0fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "da156c99",
   "metadata": {},
   "outputs": [],
   "source": [
    "class KNN:\n",
    "    def __init__(self,k = 3):\n",
    "        self.k = k\n",
    "        \n",
    "    def sample_distance(self, sample, features_train):\n",
    "        distance_2 = (sample - features_train )**2\n",
    "        dist = distance_2.sum(1)\n",
    "        dist -= np.max(dist)\n",
    "        sortedIndex = np.argsort(dist)\n",
    "        return dist, sortedIndex\n",
    "    \n",
    "    def clssifier(self,sample,features_train,label_train):\n",
    "        dist, Index = self.sample_distance(sample,features_train)\n",
    "        kIndex = Index[:self.k]\n",
    "        kPred = [label_train[Index] for Index in kIndex]\n",
    "        labels, lb_counts = np.unique(kPred,return_counts=True)\n",
    "        return labels[np.argmax(lb_counts)]\n",
    "    \n",
    "    def fit(self,features_test,features_train,label_train, progressBar=True):\n",
    "        self.pred_label = []\n",
    "        for j in range(features_test.shape[0]):\n",
    "            temp_label = self.clssifier(features_test[j,:],features_train,label_train)\n",
    "            self.pred_label.append(temp_label)\n",
    "    \n",
    "    def get_accuarcy(self,label_train):\n",
    "        print(\"accuracy\", np.mean(np.array(self.pred_label)==label_train))\n",
    "        return np.mean(np.array(self.pred_label)==label_train)\n",
    "              \n",
    "              \n",
    "#         if progressBar:\n",
    "#                 complete = (j + 1) / features_test.shape[0]\n",
    "#                 not_complete = 1 - complete\n",
    "#                 complete = int(50 * complete)\n",
    "#                 not_complete = int(50 * not_complete)\n",
    "#                 print('[' + \">\" * complete, end='')\n",
    "#                 print('-' * not_complete + ']' + '[%' + str((j + 1) * 100 / features_test.shape[0]) + ']')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "4b39edfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "8ac3c9d0",
   "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>Unnamed: 0</th>\n",
       "      <th>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  Sepal.Length  Sepal.Width  Petal.Length  Petal.Width Species\n",
       "0           1           5.1          3.5           1.4          0.2  setosa\n",
       "1           2           4.9          3.0           1.4          0.2  setosa\n",
       "2           3           4.7          3.2           1.3          0.2  setosa\n",
       "3           4           4.6          3.1           1.5          0.2  setosa\n",
       "4           5           5.0          3.6           1.4          0.2  setosa"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"iris.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "b521c943",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.drop(columns = \"Unnamed: 0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "9627fe50",
   "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>Sepal.Length</th>\n",
       "      <th>Sepal.Width</th>\n",
       "      <th>Petal.Length</th>\n",
       "      <th>Petal.Width</th>\n",
       "      <th>Species</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>setosa</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Sepal.Length  Sepal.Width  Petal.Length  Petal.Width Species\n",
       "0           5.1          3.5           1.4          0.2  setosa\n",
       "1           4.9          3.0           1.4          0.2  setosa\n",
       "2           4.7          3.2           1.3          0.2  setosa\n",
       "3           4.6          3.1           1.5          0.2  setosa\n",
       "4           5.0          3.6           1.4          0.2  setosa"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "id": "b62e5baa",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_ver = data[data['Species']=='versicolor']\n",
    "data_vir = data[data['Species']=='virginica']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "a11bb82e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.scatter(data_ver['Petal.Length'], data_ver['Petal.Width'], color='red', marker='o', label='versicolor')\n",
    "plt.scatter(data_vir['Petal.Length'], data_vir['Petal.Width'], color='blue', label='virginica')\n",
    "plt.xlabel('Petal.Length')  \n",
    "plt.ylabel('Petal.Width')  \n",
    "#plt.set_zlabel('petal length') \n",
    "plt.legend(loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "id": "26a77c80",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = np.array(pd.concat([data_ver,data_vir]))\n",
    "X = train[:,0:2]\n",
    "y = train[:,4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "id": "d24ab245",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.shuffle(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "id": "2c7442fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = train[:,0:2]\n",
    "y = train[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "id": "5cb910c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test = X[0:70,:], X[70:,:]\n",
    "y_train, y_test = y[0:70], y[70:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "98ec5e15",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNN()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "eabfa4bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn.fit(X_test,X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "96c68a31",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy 0.7\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.7"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.get_accuarcy(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "0eab2b48",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy 0.7\n",
      "accuracy 0.7\n",
      "accuracy 0.6\n",
      "accuracy 0.7\n",
      "accuracy 0.7\n",
      "accuracy 0.7\n",
      "accuracy 0.6666666666666666\n",
      "accuracy 0.6666666666666666\n",
      "accuracy 0.6666666666666666\n"
     ]
    }
   ],
   "source": [
    "# Tuning the parameter\n",
    "accuracy_ls = []\n",
    "for k in range(2,11):\n",
    "    knn = KNN(k)\n",
    "    knn.fit(X_test,X_train,y_train)\n",
    "    accuracy_ls.append(knn.get_accuarcy(y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "id": "28528493",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(range(2,11),accuracy_ls, marker='o', linestyle='-')\n",
    "plt.xlabel('k')\n",
    "plt.ylabel('Accuracy')\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}