KNN class work.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "cde62390",
   "metadata": {},
   "source": [
    "# KNN K nearest neighbours"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d13e6b44",
   "metadata": {},
   "source": [
    "## Steps followed by KNN"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2722d93",
   "metadata": {},
   "source": [
    "## Set the value of k . for example k=3 "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff56d4cc",
   "metadata": {},
   "source": [
    "## K's value will depend on number of classes \n",
    "## If no of classes is even then k will be odd\n",
    "## If no of classes is odd then k will be even"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6d471980",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e3754c24",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "de2bd907",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0e6272b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "iris = load_iris()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6009fd90",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = iris.data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "de60ea07",
   "metadata": {},
   "outputs": [],
   "source": [
    "y = iris.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bc79cc2c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150, 4)\n"
     ]
    }
   ],
   "source": [
    "print(x.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d3257bd",
   "metadata": {},
   "source": [
    "## Split data in train and test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3cec539a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0c19c6fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "xtrain,xtest,ytrain,ytest = train_test_split(x,y,test_size=.2,random_state=13)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b8b7bf6",
   "metadata": {},
   "source": [
    "## Initializing KNN class with euclidean equation metric with n_neighbors(k) = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "27e2f0f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0b4640d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(metric = 'euclidean',n_neighbors = 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61d9f905",
   "metadata": {},
   "source": [
    "## Train KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "93904886",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(metric='euclidean', n_neighbors=3)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(xtrain,ytrain) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c52fcf5c",
   "metadata": {},
   "source": [
    "## Run prediction on Trained KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c1f61b13",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred = knn.predict(xtest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "2765d255",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 0, 2, 2, 0, 2, 2, 0, 1, 2, 1, 1, 0, 2, 0, 2, 2, 1, 0, 1, 0,\n",
       "       0, 2, 1, 2, 0, 2, 1, 2])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3dd80314",
   "metadata": {},
   "source": [
    "## Testing accuracy of the Trained KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6a1a2a6c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9666666666666667"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(ytest,pred) #accuracy of the KNN model on test dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9c12b91",
   "metadata": {},
   "source": [
    "# Example 2 of KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "836e899e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import load_iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "eaf6f168",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "35c7ce74",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "9d3a8bcb",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "798c571f",
   "metadata": {},
   "outputs": [],
   "source": [
    "iris = load_iris()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ea31bcc9",
   "metadata": {},
   "outputs": [],
   "source": [
    "x1 = iris.data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "887e7014",
   "metadata": {},
   "outputs": [],
   "source": [
    "yr = iris.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "fbbf4d7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(150, 4)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x1.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16c58431",
   "metadata": {},
   "source": [
    "## Split data into train and test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9541bb63",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "66fcbaf8",
   "metadata": {},
   "outputs": [],
   "source": [
    "xtrain,xtest,ytrain,ytest = train_test_split(x,y,test_size=.2,random_state=13)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8616a7f0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(120, 4)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xtrain.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8413f05",
   "metadata": {},
   "source": [
    "## Initializing KNN class with euclidean equation metric with n_neighbors(k) = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "3fd8349a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "6668dee8",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(metric = 'euclidean')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "620036a7",
   "metadata": {},
   "source": [
    "## Train KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "a9897b70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(metric='euclidean')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(xtrain,ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "e844a905",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4,)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xtest[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "633578af",
   "metadata": {},
   "outputs": [],
   "source": [
    "t = xtest[0].reshape(1,-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "29674913",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 4)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "28fb51d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "n,m = knn.kneighbors(t,n_neighbors = 5,return_distance = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "6580b98f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 7, 12, 74, 21, 58]])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m #close neighbors index numbers from samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b04318ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.3       , 0.31622777, 0.33166248, 0.37416574, 0.4       ]])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n #distance between sample and closest neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "0147d9d4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "for i in m[0]:\n",
    "    print(ytrain[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "1b52e5ca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1]\n"
     ]
    }
   ],
   "source": [
    "print(knn.predict(t))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "0a6a745f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n"
     ]
    }
   ],
   "source": [
    "print(ytest[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "e6fae352",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 0, 2, 2, 0, 2, 2, 0, 1, 1, 1, 1, 0, 2, 0, 1, 2, 1, 0, 1, 0,\n",
       "       0, 2, 1, 2, 0, 2, 1, 2])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = knn.predict(xtest)\n",
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "3f076164",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(ytest,pred) #accuracy of the KNN model on test dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c4c9249",
   "metadata": {},
   "source": [
    "## Indexes of missclassified samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e397a2cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(array([10, 11, 16]),)\n"
     ]
    }
   ],
   "source": [
    "print(np.where(ytest!=pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0cb3b5cd",
   "metadata": {},
   "source": [
    "# Example 3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "403f2f9c",
   "metadata": {},
   "source": [
    "## Applying KNN on minset dataset "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "254b04a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f638b00",
   "metadata": {},
   "source": [
    "## downloading mnsit dataset from open_ml website"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "fc50e8b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "mnist = fetch_openml('mnist_784') # it is a dataset that contains images of numbers from 0 to 9"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ce1f65ff",
   "metadata": {},
   "source": [
    "## x and y comes in the form of dataframe hence use .values to convert them into arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d192d69e",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_num = mnist.data.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "49a91986",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_num = mnist.target.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "65a4e7bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000, 784)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_num.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "a78c6476",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_num.ndim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "10b9d46c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(70000,)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_num.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "d70b378a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_num.ndim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "394964d4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_num"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "772e09de",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt #importing pyplot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "a5ec7d44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x_num[2].reshape(28,28),cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "b191dc8b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array(['0', '1', '2', '3', '4', '5', '6', '7', '8', '9'], dtype=object),\n",
       " array([6903, 7877, 6990, 7141, 6824, 6313, 6876, 7293, 6825, 6958]))"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(y_num,return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "824a929c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAN5ElEQVR4nO3df6hc9ZnH8c+nsQFNFZMVs8HGtS0imxg2XSQuRtasWjH+YVR0rQRx3bK3YAMWFl3N/lF1rdRl20WEFK4oxqVrEbUaTWMbTVndP6wmwY1Js9Yobhu95pdgLag1ybN/3JNyjfd85zozZ86Y5/2Cy8yc555zngz53HNmvjPn64gQgCPf59puAMBgEHYgCcIOJEHYgSQIO5DEUYPcmW3e+gcaFhGebHlPR3bbF9p+xfYO2zf1si0AzXK34+y2p0n6taSvSdop6UVJV0XErwrrcGQHGtbEkX2RpB0R8XpE/EHSjyUt62F7ABrUS9hPkvTbCY93Vss+xvaI7Y22N/awLwA96uUNuslOFT5xmh4Ro5JGJU7jgTb1cmTfKWnuhMdflPRWb+0AaEovYX9R0qm2v2R7uqSvS1rTn7YA9FvXp/ERsd/2Ckk/kzRN0n0Rsa1vnQHoq66H3rraGa/ZgcY18qEaAJ8dhB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSRB2IAnCDiTR9ZTN6J+5c+cW6zfffHOxvmDBgtra4sWLi+vak074+UedZvlds2ZNsT5z5sza2rZt5Rm+X3jhhWL9/vvvL9bxcT2F3fYbkt6TdEDS/og4ox9NAei/fhzZ/yYi9vZhOwAaxGt2IIlewx6Sfm57k+2RyX7B9ojtjbY39rgvAD3o9TR+cUS8ZftESett/29EPDvxFyJiVNKoJNkuv9sDoDE9Hdkj4q3qdrekn0ha1I+mAPRf12G3PcP2sYfuS7pA0tZ+NQagv9xpHLV2RfvLGj+aS+MvB/4zIr7bYZ0j8jR+4cKFxfqNN95YrJ911lnFeqdx+JJ9+/YV66+88kqx3qm3Ju3dWx7kmT179oA6+WyJiEk/PNH1a/aIeF3SX3TdEYCBYugNSIKwA0kQdiAJwg4kQdiBJPiKa+Xqq68u1letWlVbmz59enHdo44qP80bNmwo1i+++OJifceOHbW1gwcPFtfdv39/sd7p3/bUU08V652+YovB4cgOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kwzl457rjjivVjjjmm623v2rWrWL/hhhuK9S1btnS97151GofvNI7fiyeeeKKxbWfEkR1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkuj6UtJd7WyILyU9bdq0Yr009XAnH330UbH+7rvvdr3tps2fP79YX7t2bbFeugz2Bx98UFz38ssvL9bXrVtXrGdVdylpjuxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kATj7Cj68MMPi/VO18QvjaXfeeedxXVvu+22Yh2T63qc3fZ9tnfb3jph2Szb622/Wt12/4kTAAMxldP4+yVdeNiymyQ9ExGnSnqmegxgiHUMe0Q8K+mdwxYvk7S6ur9a0iX9bQtAv3V7DbrZETEmSRExZvvEul+0PSJppMv9AOiTxi84GRGjkkYl3qAD2tTt0Nsu23Mkqbrd3b+WADSh27CvkXRNdf8aSY/3px0ATek4zm77QUlLJJ0gaZek70h6TNJDkk6W9BtJV0TE4W/iTbYtTuMbULrm/ZVXXllcd+XKlcX6ySefXKx3+q7+7bff3lUN3asbZ+/4mj0irqopnddTRwAGio/LAkkQdiAJwg4kQdiBJAg7kARTNg+BGTNmFOv33HNPsb506dLaWqepqHv13HPPFesPPPBAo/vH1HFkB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkuJT0EDj++OOL9bfffrtY/9zn6v9md5qKuml79uypre3bt6+47ujoaLF+9913F+sHDx4s1o9UTNkMJEfYgSQIO5AEYQeSIOxAEoQdSIKwA0kwzn4EmD9/fm1t0aJFPW37+uuvL9YXLFjQ0/Z7sWHDhmJ9+fLltbXdu4/ceU0YZweSI+xAEoQdSIKwA0kQdiAJwg4kQdiBJBhnR9HRRx9drM+bN69YP//882trd9xxR1c9TdWyZctqa08++WSj+25T1+Pstu+zvdv21gnLbrH9pu2Xqp+L+tksgP6bymn8/ZIunGT5v0fEwurnp/1tC0C/dQx7RDwr6Z0B9AKgQb28QbfC9pbqNH9m3S/ZHrG90fbGHvYFoEfdhv2Hkr4iaaGkMUnfr/vFiBiNiDMi4owu9wWgD7oKe0TsiogDEXFQ0j2SevtqFYDGdRV223MmPLxU0ta63wUwHDrOz277QUlLJJ1ge6ek70haYnuhpJD0hqRvNtci2vT+++8X65s2bSrWN2/eXFtbsmRJcd0LLrigWO/knHPOqa0dyePsdTqGPSKummTxvQ30AqBBfFwWSIKwA0kQdiAJwg4kQdiBJDq+Gw/0ovQV6qa/Xv3aa681uv3PGo7sQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AE4+xo1BVXXFFbO++88xrd99NPP93o9j9rOLIDSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKMs6MnZ599drF+66231taOOqq3/36PPfZYsT42NtbT9o80HNmBJAg7kARhB5Ig7EAShB1IgrADSRB2IAk3fe3uj+3MHtzO0BfXXnttsb5q1apiffr06V3v+8033yzWTzvttGK903TTR6qI8GTLOx7Zbc+1/Qvb221vs319tXyW7fW2X61uZ/a7aQD9M5XT+P2S/jEi/lzSX0n6lu15km6S9ExEnCrpmeoxgCHVMewRMRYRm6v770naLukkScskra5+bbWkSxrqEUAffKoPJ9s+RdJXJf1S0uyIGJPG/yDYPrFmnRFJIz32CaBHUw677S9IekTStyPid/ak7wF8QkSMShqttsEbdEBLpjT0ZvvzGg/6jyLi0WrxLttzqvocSbubaRFAP3Q8snv8EH6vpO0R8YMJpTWSrpH0ver28UY6RE/mzZtXrK9YsaJYHxkpvwKb6hneZPbu3VusX3bZZcV61qG1bk3lNH6xpKslvWz7pWrZSo2H/CHb35D0G0n1FwgH0LqOYY+I/5ZU9+e72av8A+gbPi4LJEHYgSQIO5AEYQeSIOxAElxKeopK49VLly4trrtu3bpifdasWcX6mWeeWayffvrptbVLL720uO6xxx5brHdy4MCBYn3t2rW1teuuu664LpeC7i+O7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBJeSnqL169fX1s4999wBdjJYzz//fLF+1113FesPPfRQP9vBFHR9KWkARwbCDiRB2IEkCDuQBGEHkiDsQBKEHUiC77NP0cMPP1xbG+Zx9j179hTry5cvL9Y3bNhQrA/ycxroDUd2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUii4/fZbc+V9ICkP5V0UNJoRNxl+xZJ/yDp0EDuyoj4aYdtMSgLNKzu++xTCfscSXMiYrPtYyVtknSJpL+V9PuI+LepNkHYgebVhX0q87OPSRqr7r9ne7ukk/rbHoCmfarX7LZPkfRVSb+sFq2wvcX2fbZn1qwzYnuj7Y29tQqgF1O+Bp3tL0j6L0nfjYhHbc+WtFdSSPoXjZ/q/32HbXAaDzSs69fskmT785KelPSziPjBJPVTJD0ZEfUzDIqwA4PQ9QUnbVvSvZK2Twx69cbdIZdK2tprkwCaM5V348+W9JyklzU+9CZJKyVdJWmhxk/j35D0zerNvNK2OLIDDevpNL5fCDvQPK4bDyRH2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSGLQUzbvlfR/Ex6fUC0bRsPa27D2JdFbt/rZ25/VFQb6ffZP7NzeGBFntNZAwbD2Nqx9SfTWrUH1xmk8kARhB5JoO+yjLe+/ZFh7G9a+JHrr1kB6a/U1O4DBafvIDmBACDuQRCtht32h7Vds77B9Uxs91LH9hu2Xbb/U9vx01Rx6u21vnbBslu31tl+tbiedY6+l3m6x/Wb13L1k+6KWeptr+xe2t9veZvv6anmrz12hr4E8bwN/zW57mqRfS/qapJ2SXpR0VUT8aqCN1LD9hqQzIqL1D2DY/mtJv5f0wKGptWz/q6R3IuJ71R/KmRHxT0PS2y36lNN4N9Rb3TTjf6cWn7t+Tn/ejTaO7Isk7YiI1yPiD5J+LGlZC30MvYh4VtI7hy1eJml1dX+1xv+zDFxNb0MhIsYiYnN1/z1Jh6YZb/W5K/Q1EG2E/SRJv53weKeGa773kPRz25tsj7TdzCRmH5pmq7o9seV+DtdxGu9BOmya8aF57rqZ/rxXbYR9sqlphmn8b3FE/KWkpZK+VZ2uYmp+KOkrGp8DcEzS99tspppm/BFJ346I37XZy0ST9DWQ562NsO+UNHfC4y9KequFPiYVEW9Vt7sl/UTjLzuGya5DM+hWt7tb7uePImJXRByIiIOS7lGLz101zfgjkn4UEY9Wi1t/7ibra1DPWxthf1HSqba/ZHu6pK9LWtNCH59ge0b1xolsz5B0gYZvKuo1kq6p7l8j6fEWe/mYYZnGu26acbX83LU+/XlEDPxH0kUaf0f+NUn/3EYPNX19WdL/VD/b2u5N0oMaP637SONnRN+Q9CeSnpH0anU7a4h6+w+NT+29RePBmtNSb2dr/KXhFkkvVT8Xtf3cFfoayPPGx2WBJPgEHZAEYQeSIOxAEoQdSIKwA0kQdiAJwg4k8f87plsS5KGGsAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x_num[10000].reshape(28,28),cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "043c6611",
   "metadata": {},
   "source": [
    "## Split data into train and test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "38601444",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "0b06939f",
   "metadata": {},
   "outputs": [],
   "source": [
    "xtrain_num,xtest_num,ytrain_num,ytest_num = train_test_split(x_num,y_num,test_size=.2,random_state=13)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "086be0b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "knn = KNeighborsClassifier(metric = 'euclidean')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "560b5ff6",
   "metadata": {},
   "source": [
    "## Train KNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "b1c259a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(metric='euclidean')"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn.fit(xtrain_num,ytrain_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "e735e900",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xtest_num[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "cda9870c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xtest_num[0].ndim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "48d53412",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = xtest_num[0].reshape(1,-1) # here we are changing xtest[0] from 1D array into 2D array\n",
    "t.ndim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "8c77dcd1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 784)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "d4b6653d",
   "metadata": {},
   "outputs": [],
   "source": [
    "n,m = knn.kneighbors(t,n_neighbors = 5,return_distance = True) # this only excepts a 2D numpy array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "a0ca01c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[30768, 45364, 35911, 28480, 36280]])"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m #close neighbors index numbers from samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "e74be575",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[858.77179739, 929.84568612, 931.12136695, 983.39615619,\n",
       "        996.11143955]])"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "n #distance between sample and closest neighbors"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94ab8bbd",
   "metadata": {},
   "source": [
    "## Checking the values in ytrain of m[0] array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "a3becc33",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n",
      "7\n",
      "7\n",
      "7\n",
      "7\n"
     ]
    }
   ],
   "source": [
    "for i in m[0]:\n",
    "    print(ytrain_num[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a741b75",
   "metadata": {},
   "source": [
    "## Testing predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "8428c86d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['7']\n"
     ]
    }
   ],
   "source": [
    "print(knn.predict(t))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "299e9531",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7\n"
     ]
    }
   ],
   "source": [
    "print(ytest_num[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56e51cbf",
   "metadata": {},
   "source": [
    "## Running predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "bd8430ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['7', '3', '8', ..., '2', '2', '4'], dtype=object)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = knn.predict(xtest_num)\n",
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "6cfdda02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['7', '3', '8', '6', '6', ..., '9', '4', '2', '2', '4']\n",
       "Length: 14000\n",
       "Categories (10, object): ['0', '1', '2', '3', ..., '6', '7', '8', '9']"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ytest_num"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3078e5e0",
   "metadata": {},
   "source": [
    "## Testing accuracy of the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "2b0fd77f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.97"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(ytest_num,pred) #accuracy of the KNN model on test dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05ad0f96",
   "metadata": {},
   "source": [
    "## Getting misclassified samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "f11e79df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(array([   36,    60,    71,   143,   162,   226,   235,   250,   253,\n",
      "         256,   333,   350,   379,   477,   583,   596,   604,   695,\n",
      "         713,   718,   776,   810,   811,   842,   937,   982,   989,\n",
      "         991,  1096,  1103,  1132,  1255,  1306,  1345,  1400,  1433,\n",
      "        1447,  1478,  1563,  1579,  1582,  1620,  1650,  1655,  1662,\n",
      "        1758,  1833,  1850,  1875,  1890,  2002,  2026,  2048,  2087,\n",
      "        2096,  2186,  2188,  2208,  2218,  2222,  2275,  2312,  2379,\n",
      "        2423,  2428,  2430,  2432,  2466,  2499,  2562,  2672,  2690,\n",
      "        2704,  2744,  2764,  2794,  2828,  2899,  2923,  2929,  2931,\n",
      "        2932,  2974,  3049,  3073,  3080,  3115,  3164,  3192,  3257,\n",
      "        3266,  3286,  3296,  3308,  3339,  3410,  3460,  3480,  3494,\n",
      "        3634,  3661,  3673,  3682,  3693,  3713,  3733,  3825,  3834,\n",
      "        3857,  3881,  3912,  4037,  4079,  4120,  4122,  4170,  4206,\n",
      "        4212,  4278,  4287,  4298,  4299,  4300,  4351,  4365,  4368,\n",
      "        4424,  4429,  4622,  4632,  4638,  4645,  4703,  4765,  4771,\n",
      "        4807,  4854,  4951,  4955,  4974,  4980,  5076,  5111,  5135,\n",
      "        5137,  5167,  5182,  5185,  5224,  5258,  5304,  5318,  5325,\n",
      "        5337,  5360,  5373,  5401,  5408,  5412,  5422,  5429,  5481,\n",
      "        5490,  5542,  5546,  5702,  5825,  5830,  5927,  5953,  5963,\n",
      "        5966,  5971,  5985,  5992,  6010,  6079,  6088,  6138,  6139,\n",
      "        6151,  6172,  6263,  6287,  6305,  6314,  6347,  6369,  6382,\n",
      "        6399,  6420,  6437,  6512,  6536,  6551,  6555,  6561,  6592,\n",
      "        6615,  6627,  6670,  6703,  6733,  6738,  6864,  6891,  6899,\n",
      "        6910,  6999,  7186,  7190,  7221,  7250,  7263,  7302,  7338,\n",
      "        7438,  7469,  7534,  7555,  7576,  7590,  7597,  7625,  7659,\n",
      "        7662,  7671,  7712,  7762,  7812,  7860,  7882,  7936,  7953,\n",
      "        7989,  8002,  8010,  8056,  8115,  8139,  8149,  8216,  8274,\n",
      "        8277,  8366,  8367,  8416,  8419,  8443,  8477,  8508,  8586,\n",
      "        8632,  8652,  8685,  8755,  8767,  8928,  8943,  8998,  9026,\n",
      "        9035,  9049,  9061,  9065,  9137,  9189,  9193,  9200,  9255,\n",
      "        9278,  9287,  9315,  9419,  9440,  9454,  9471,  9565,  9598,\n",
      "        9610,  9623,  9643,  9645,  9648,  9773,  9807,  9848,  9865,\n",
      "        9868,  9891,  9952, 10030, 10086, 10096, 10112, 10159, 10170,\n",
      "       10194, 10228, 10240, 10248, 10346, 10354, 10405, 10419, 10456,\n",
      "       10506, 10516, 10531, 10558, 10578, 10628, 10656, 10658, 10691,\n",
      "       10696, 10704, 10707, 10760, 10770, 10789, 10820, 10867, 10880,\n",
      "       10917, 10955, 10999, 11013, 11021, 11068, 11110, 11128, 11137,\n",
      "       11139, 11318, 11353, 11447, 11457, 11477, 11505, 11527, 11533,\n",
      "       11536, 11541, 11577, 11598, 11644, 11645, 11696, 11705, 11768,\n",
      "       11825, 11842, 11899, 11944, 11973, 11996, 12002, 12076, 12108,\n",
      "       12109, 12159, 12209, 12219, 12325, 12402, 12422, 12466, 12499,\n",
      "       12502, 12513, 12535, 12537, 12547, 12563, 12597, 12614, 12622,\n",
      "       12666, 12684, 12702, 12707, 12710, 12711, 12749, 12827, 12838,\n",
      "       12840, 12855, 12856, 12877, 12920, 12964, 12980, 12989, 13111,\n",
      "       13114, 13148, 13159, 13232, 13319, 13333, 13370, 13423, 13468,\n",
      "       13472, 13473, 13486, 13514, 13517, 13524, 13537, 13551, 13569,\n",
      "       13580, 13594, 13811, 13822, 13900, 13996]),)\n"
     ]
    }
   ],
   "source": [
    "print(np.where(ytest_num!=pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5fd4cbfa",
   "metadata": {},
   "source": [
    "## Now importing our cutom image and try to classify the number using KNN"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57f7f74a",
   "metadata": {},
   "source": [
    "## Importing image in this jupyter notebook using OpenCv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "d25a025b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "5ec8bcf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing image\n",
    "image = cv2.imread('/home/Machine_learning/github/Machine_Learning/Prutor_ML_ClassWork/six.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "ab37ffd3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image.ndim #here image imported via opencv is in 3D numpy array  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "56906d0d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f1084164550>"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(image)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25bcc1ce",
   "metadata": {},
   "source": [
    "## Because we trained our model using numbers of color white with a black background we need to invert the color of our image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "54f9d496",
   "metadata": {},
   "outputs": [],
   "source": [
    "image1 = np.invert(image) #inverting color using numpy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "d6ee55dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f10b9c897c0>"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(image1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "342dc878",
   "metadata": {},
   "outputs": [],
   "source": [
    "gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "6c6d39dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gray.ndim #it will convert 3D array of image into 2D array of image "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aac8971a",
   "metadata": {},
   "source": [
    "## We trained our model using 28 x 28 size images so we need to resize our imported image in 28 x 28 size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "6b4fd5a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "resized_img = cv2.resize(gray,(28,28)) #resize image into 28 x 28"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "0cc17f28",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(28, 28)"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resized_img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "6a11a26b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resized_img.ndim"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bc4b683",
   "metadata": {},
   "source": [
    "## reshaping image because our KNN model is expecting 784 features in the input data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "677ac8ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "reshaped_image = resized_img.reshape(1,784) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "9195d991",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reshaped_image.ndim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "ccb1a42a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['7'], dtype=object)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = knn.predict(reshaped_image)\n",
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "892362b8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# try using minmax scalar here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1ed94df4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:tensorflow] *",
   "language": "python",
   "name": "conda-env-tensorflow-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}