{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "38b3b8f6", "metadata": { "scrolled": true }, "outputs": [], "source": [ "# 1. We need to include all of these packages to make sure \n", "# that all of the things we can to use can be.\n", "from sklearn import neighbors, datasets\n", "from sklearn.neighbors import KNeighborsClassifier\n", "from sklearn.model_selection import train_test_split\n", "from sklearn import metrics\n", "import numpy as np\n", "import pandas as pd\n", "\n", "# We then need to load in the datasets \n", "mnistTrain = pd.read_csv('fashion-mnist_train.csv')" ] }, { "cell_type": "code", "execution_count": 2, "id": "976c52dc", "metadata": { "scrolled": true }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " label pixel1 pixel2 pixel3 pixel4 pixel5 pixel6 pixel7 pixel8 \\\n", "0 2 0 0 0 0 0 0 0 0 \n", "1 9 0 0 0 0 0 0 0 0 \n", "2 6 0 0 0 0 0 0 0 5 \n", "3 0 0 0 0 1 2 0 0 0 \n", "4 3 0 0 0 0 0 0 0 0 \n", "\n", " pixel9 ... pixel775 pixel776 pixel777 pixel778 pixel779 pixel780 \\\n", "0 0 ... 0 0 0 0 0 0 \n", "1 0 ... 0 0 0 0 0 0 \n", "2 0 ... 0 0 0 30 43 0 \n", "3 0 ... 3 0 0 0 0 1 \n", "4 0 ... 0 0 0 0 0 0 \n", "\n", " pixel781 pixel782 pixel783 pixel784 \n", "0 0 0 0 0 \n", "1 0 0 0 0 \n", "2 0 0 0 0 \n", "3 0 0 0 0 \n", "4 0 0 0 0 \n", "\n", "[5 rows x 785 columns]\n" ] } ], "source": [ "# 2. We can view that this data looks like by using the following code.\n", "print(mnistTrain.head())" ] }, { "cell_type": "code", "execution_count": 3, "id": "bdfd167e", "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training data points: 53730\n", "Validation data points: 5970\n", "Testing data points: 300\n" ] }, { "data": { "text/html": [ "
KNeighborsClassifier(n_neighbors=10)
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" ], "text/plain": [ "KNeighborsClassifier(n_neighbors=10)" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 3. We now need to take the data that we have which is the MNISTFashion \n", "# data called 'mnistTrain'. We will split it up 75% for training and \n", "# 25% for testing.\n", "(trainData, testData, trainLabels, testLabels) = train_test_split(np.array(\n", " mnistTrain),mnistTrain, test_size=0.005, random_state=42)\n", "\n", "#(trainData, testData, trainLabels, testLabels) = train_test_split(np.array(\n", "# mnistTrain),mnistTrain, test_size=0.25, random_state=42)\n", "\n", "# now, let's take 10% of the training data and use that for validation\n", "(trainData, valData, trainLabels, valLabels) = train_test_split(\n", " trainData, trainLabels,test_size=0.1, random_state=84)\n", "\n", "# Now we can see the amount that have been split.\n", "print(\"Training data points: {}\".format(len(trainLabels)))\n", "print(\"Validation data points: {}\".format(len(valLabels)))\n", "print(\"Testing data points: {}\".format(len(testLabels)))\n", "\n", "model = neighbors.KNeighborsClassifier(n_neighbors=10)\n", "model.fit(trainData, trainLabels)" ] }, { "cell_type": "code", "execution_count": 4, "id": "be100307", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train Data Shape: (53730, 785)\n" ] } ], "source": [ "# 4.Finding out the train data shape\n", "print(\"Train Data Shape:\", trainData.shape)" ] }, { "cell_type": "code", "execution_count": 5, "id": "403cc94d", "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "array([[7, 0, 0, ..., 0, 0, 0],\n", " [8, 0, 0, ..., 0, 0, 0],\n", " [8, 0, 0, ..., 0, 0, 0],\n", " ...,\n", " [8, 0, 0, ..., 0, 0, 0],\n", " [2, 0, 0, ..., 0, 0, 0],\n", " [8, 0, 0, ..., 0, 0, 0]], dtype=int64)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 5. Now we can make a prediction using the test data \n", "y_pred2=model.predict(testData)\n", "y_pred2" ] }, { "cell_type": "code", "execution_count": 6, "id": "57689936", "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
labelpixel1pixel2pixel3pixel4pixel5pixel6pixel7pixel8pixel9...pixel775pixel776pixel777pixel778pixel779pixel780pixel781pixel782pixel783pixel784
126287000000000...0000000000
377308000000000...1871901962012092212182111020
399918000000000...0000000000
85255000000000...0100000000
82799000000000...0000000000
..................................................................
29804000000000...209223218330000
546120000000000...511100000000
496868000000000...0000000000
310672000010000...020213811461000
514028000000000...0001000000
\n", "

300 rows × 785 columns

\n", "
" ], "text/plain": [ " label pixel1 pixel2 pixel3 pixel4 pixel5 pixel6 pixel7 pixel8 \\\n", "12628 7 0 0 0 0 0 0 0 0 \n", "37730 8 0 0 0 0 0 0 0 0 \n", "39991 8 0 0 0 0 0 0 0 0 \n", "8525 5 0 0 0 0 0 0 0 0 \n", "8279 9 0 0 0 0 0 0 0 0 \n", "... ... ... ... ... ... ... ... ... ... \n", "2980 4 0 0 0 0 0 0 0 0 \n", "54612 0 0 0 0 0 0 0 0 0 \n", "49686 8 0 0 0 0 0 0 0 0 \n", "31067 2 0 0 0 0 1 0 0 0 \n", "51402 8 0 0 0 0 0 0 0 0 \n", "\n", " pixel9 ... pixel775 pixel776 pixel777 pixel778 pixel779 \\\n", "12628 0 ... 0 0 0 0 0 \n", "37730 0 ... 187 190 196 201 209 \n", "39991 0 ... 0 0 0 0 0 \n", "8525 0 ... 0 1 0 0 0 \n", "8279 0 ... 0 0 0 0 0 \n", "... ... ... ... ... ... ... ... \n", "2980 0 ... 2 0 9 223 218 \n", "54612 0 ... 51 11 0 0 0 \n", "49686 0 ... 0 0 0 0 0 \n", "31067 0 ... 0 2 0 2 138 \n", "51402 0 ... 0 0 0 1 0 \n", "\n", " pixel780 pixel781 pixel782 pixel783 pixel784 \n", "12628 0 0 0 0 0 \n", "37730 221 218 211 102 0 \n", "39991 0 0 0 0 0 \n", "8525 0 0 0 0 0 \n", "8279 0 0 0 0 0 \n", "... ... ... ... ... ... \n", "2980 33 0 0 0 0 \n", "54612 0 0 0 0 0 \n", "49686 0 0 0 0 0 \n", "31067 114 61 0 0 0 \n", "51402 0 0 0 0 0 \n", "\n", "[300 rows x 785 columns]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#6. Now we can check the test labels \n", "testLabels" ] }, { "cell_type": "code", "execution_count": 7, "id": "d3942e63", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[114600 113 52 ... 0 0 7]\n", " [ 2012 99 25 ... 0 0 0]\n", " [ 1183 43 64 ... 0 0 0]\n", " ...\n", " [ 13 0 0 ... 4 5 43]\n", " [ 14 0 0 ... 3 1 71]\n", " [ 101 2 3 ... 18 28 668]]\n" ] } ], "source": [ "# 7. Now we are going to make a confusion matrix\n", "print(metrics.confusion_matrix(testLabels.to_numpy().ravel(), \n", " y_pred2.ravel()))\n", "\n" ] }, { "cell_type": "code", "execution_count": 8, "id": "9af99126", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " 0 0.834 0.992 0.906 115552\n", " 1 0.171 0.045 0.071 2207\n", " 2 0.159 0.047 0.072 1374\n", " 3 0.153 0.045 0.070 990\n", " 4 0.102 0.034 0.051 700\n", " 5 0.107 0.039 0.058 533\n", " 6 0.110 0.053 0.072 433\n", " 7 0.124 0.071 0.090 325\n", " 8 0.161 0.104 0.126 289\n", " 9 0.173 0.135 0.152 252\n", " 10 0.000 0.000 0.000 220\n", " 11 0.006 0.005 0.005 191\n", " 12 0.011 0.011 0.011 182\n", " 13 0.028 0.026 0.027 196\n", " 14 0.006 0.004 0.005 225\n", " 15 0.006 0.006 0.006 180\n", " 16 0.015 0.016 0.016 182\n", " 17 0.005 0.005 0.005 216\n", " 18 0.000 0.000 0.000 212\n", " 19 0.005 0.004 0.005 225\n", " 20 0.010 0.008 0.009 260\n", " 21 0.005 0.005 0.005 222\n", " 22 0.012 0.015 0.014 197\n", " 23 0.017 0.014 0.016 277\n", " 24 0.010 0.008 0.009 237\n", " 25 0.016 0.013 0.014 300\n", " 26 0.008 0.009 0.008 212\n", " 27 0.008 0.007 0.007 282\n", " 28 0.008 0.007 0.007 303\n", " 29 0.027 0.030 0.028 271\n", " 30 0.004 0.004 0.004 258\n", " 31 0.000 0.000 0.000 199\n", " 32 0.021 0.020 0.021 293\n", " 33 0.009 0.008 0.008 258\n", " 34 0.016 0.015 0.015 274\n", " 35 0.004 0.004 0.004 270\n", " 36 0.014 0.012 0.013 347\n", " 37 0.004 0.005 0.004 217\n", " 38 0.004 0.003 0.004 286\n", " 39 0.007 0.007 0.007 302\n", " 40 0.018 0.015 0.017 334\n", " 41 0.007 0.007 0.007 294\n", " 42 0.004 0.005 0.004 207\n", " 43 0.015 0.013 0.014 302\n", " 44 0.003 0.003 0.003 313\n", " 45 0.013 0.017 0.015 232\n", " 46 0.007 0.008 0.008 251\n", " 47 0.007 0.006 0.007 335\n", " 48 0.000 0.000 0.000 279\n", " 49 0.011 0.012 0.011 256\n", " 50 0.004 0.003 0.003 292\n", " 51 0.010 0.011 0.010 283\n", " 52 0.016 0.016 0.016 311\n", " 53 0.003 0.004 0.003 279\n", " 54 0.007 0.006 0.006 325\n", " 55 0.016 0.016 0.016 311\n", " 56 0.014 0.012 0.013 333\n", " 57 0.000 0.000 0.000 230\n", " 58 0.003 0.003 0.003 333\n", " 59 0.000 0.000 0.000 281\n", " 60 0.011 0.015 0.013 269\n", " 61 0.010 0.010 0.010 292\n", " 62 0.006 0.007 0.007 300\n", " 63 0.004 0.004 0.004 256\n", " 64 0.009 0.011 0.010 280\n", " 65 0.003 0.004 0.003 256\n", " 66 0.016 0.016 0.016 307\n", " 67 0.009 0.007 0.008 290\n", " 68 0.009 0.010 0.010 298\n", " 69 0.021 0.022 0.022 314\n", " 70 0.020 0.021 0.020 339\n", " 71 0.000 0.000 0.000 256\n", " 72 0.021 0.018 0.020 380\n", " 73 0.000 0.000 0.000 214\n", " 74 0.004 0.003 0.003 305\n", " 75 0.009 0.009 0.009 318\n", " 76 0.021 0.021 0.021 284\n", " 77 0.000 0.000 0.000 219\n", " 78 0.022 0.019 0.020 318\n", " 79 0.003 0.004 0.004 236\n", " 80 0.007 0.006 0.006 326\n", " 81 0.000 0.000 0.000 268\n", " 82 0.007 0.006 0.006 346\n", " 83 0.009 0.011 0.010 264\n", " 84 0.009 0.008 0.008 253\n", " 85 0.010 0.012 0.011 338\n", " 86 0.003 0.003 0.003 322\n", " 87 0.003 0.004 0.003 258\n", " 88 0.006 0.007 0.007 284\n", " 89 0.003 0.003 0.003 322\n", " 90 0.012 0.014 0.013 282\n", " 91 0.003 0.004 0.003 270\n", " 92 0.000 0.000 0.000 319\n", " 93 0.003 0.004 0.003 277\n", " 94 0.009 0.010 0.009 306\n", " 95 0.005 0.008 0.006 266\n", " 96 0.009 0.008 0.008 372\n", " 97 0.003 0.003 0.003 286\n", " 98 0.006 0.009 0.007 235\n", " 99 0.003 0.003 0.003 315\n", " 100 0.003 0.003 0.003 298\n", " 101 0.004 0.004 0.004 266\n", " 102 0.005 0.006 0.005 358\n", " 103 0.011 0.013 0.012 389\n", " 104 0.003 0.003 0.003 335\n", " 105 0.009 0.011 0.010 369\n", " 106 0.003 0.004 0.003 284\n", " 107 0.000 0.000 0.000 351\n", " 108 0.007 0.008 0.007 358\n", " 109 0.002 0.003 0.003 298\n", " 110 0.011 0.010 0.010 401\n", " 111 0.005 0.006 0.005 308\n", " 112 0.011 0.015 0.013 324\n", " 113 0.005 0.006 0.005 362\n", " 114 0.002 0.003 0.003 308\n", " 115 0.008 0.009 0.008 348\n", " 116 0.004 0.006 0.005 341\n", " 117 0.009 0.011 0.010 348\n", " 118 0.002 0.003 0.003 339\n", " 119 0.006 0.008 0.007 381\n", " 120 0.000 0.000 0.000 388\n", " 121 0.005 0.005 0.005 384\n", " 122 0.000 0.000 0.000 418\n", " 123 0.002 0.003 0.003 341\n", " 124 0.007 0.007 0.007 401\n", " 125 0.012 0.012 0.012 412\n", " 126 0.006 0.008 0.007 391\n", " 127 0.000 0.000 0.000 314\n", " 128 0.005 0.005 0.005 411\n", " 129 0.005 0.005 0.005 387\n", " 130 0.004 0.004 0.004 479\n", " 131 0.002 0.003 0.003 368\n", " 132 0.000 0.000 0.000 401\n", " 133 0.004 0.005 0.005 366\n", " 134 0.003 0.003 0.003 379\n", " 135 0.007 0.007 0.007 407\n", " 136 0.009 0.010 0.009 408\n", " 137 0.002 0.002 0.002 438\n", " 138 0.002 0.002 0.002 422\n", " 139 0.004 0.005 0.005 389\n", " 140 0.016 0.020 0.017 405\n", " 141 0.002 0.002 0.002 500\n", " 142 0.002 0.002 0.002 417\n", " 143 0.000 0.000 0.000 443\n", " 144 0.006 0.006 0.006 506\n", " 145 0.009 0.011 0.010 455\n", " 146 0.008 0.010 0.009 405\n", " 147 0.002 0.002 0.002 452\n", " 148 0.006 0.007 0.006 454\n", " 149 0.006 0.006 0.006 483\n", " 150 0.002 0.002 0.002 453\n", " 151 0.008 0.009 0.009 453\n", " 152 0.005 0.005 0.005 433\n", " 153 0.005 0.006 0.006 492\n", " 154 0.008 0.010 0.009 401\n", " 155 0.002 0.002 0.002 529\n", " 156 0.012 0.013 0.013 448\n", " 157 0.019 0.018 0.019 545\n", " 158 0.013 0.012 0.013 569\n", " 159 0.008 0.008 0.008 502\n", " 160 0.011 0.011 0.011 566\n", " 161 0.007 0.006 0.006 540\n", " 162 0.017 0.016 0.017 556\n", " 163 0.013 0.011 0.011 571\n", " 164 0.007 0.009 0.008 465\n", " 165 0.008 0.007 0.007 581\n", " 166 0.006 0.005 0.006 571\n", " 167 0.008 0.007 0.007 578\n", " 168 0.010 0.011 0.010 473\n", " 169 0.005 0.005 0.005 427\n", " 170 0.005 0.005 0.005 638\n", " 171 0.012 0.011 0.011 565\n", " 172 0.007 0.006 0.006 684\n", " 173 0.010 0.009 0.009 672\n", " 174 0.007 0.006 0.007 631\n", " 175 0.004 0.003 0.003 690\n", " 176 0.011 0.010 0.010 624\n", " 177 0.011 0.009 0.010 640\n", " 178 0.012 0.010 0.011 711\n", " 179 0.007 0.006 0.007 620\n", " 180 0.005 0.005 0.005 635\n", " 181 0.004 0.005 0.005 620\n", " 182 0.010 0.009 0.009 659\n", " 183 0.011 0.011 0.011 623\n", " 184 0.013 0.012 0.012 688\n", " 185 0.016 0.015 0.016 659\n", " 186 0.012 0.012 0.012 686\n", " 187 0.013 0.010 0.011 769\n", " 188 0.006 0.006 0.006 698\n", " 189 0.010 0.009 0.009 666\n", " 190 0.022 0.020 0.021 738\n", " 191 0.006 0.005 0.005 640\n", " 192 0.011 0.010 0.010 736\n", " 193 0.009 0.007 0.008 701\n", " 194 0.013 0.011 0.012 731\n", " 195 0.007 0.005 0.006 813\n", " 196 0.007 0.005 0.006 766\n", " 197 0.014 0.011 0.012 753\n", " 198 0.012 0.009 0.010 764\n", " 199 0.016 0.013 0.014 782\n", " 200 0.017 0.013 0.015 792\n", " 201 0.018 0.014 0.016 712\n", " 202 0.013 0.010 0.012 781\n", " 203 0.018 0.011 0.014 786\n", " 204 0.021 0.016 0.018 795\n", " 205 0.013 0.009 0.010 822\n", " 206 0.006 0.005 0.005 842\n", " 207 0.005 0.004 0.004 778\n", " 208 0.008 0.006 0.007 887\n", " 209 0.013 0.009 0.011 843\n", " 210 0.015 0.011 0.012 854\n", " 211 0.013 0.008 0.010 853\n", " 212 0.030 0.021 0.025 803\n", " 213 0.025 0.015 0.019 885\n", " 214 0.008 0.005 0.006 842\n", " 215 0.017 0.009 0.011 913\n", " 216 0.033 0.018 0.023 946\n", " 217 0.023 0.013 0.017 925\n", " 218 0.030 0.018 0.022 910\n", " 219 0.026 0.012 0.017 885\n", " 220 0.015 0.009 0.011 798\n", " 221 0.016 0.008 0.011 841\n", " 222 0.037 0.018 0.024 850\n", " 223 0.024 0.012 0.016 737\n", " 224 0.030 0.013 0.018 854\n", " 225 0.033 0.015 0.021 844\n", " 226 0.010 0.005 0.007 821\n", " 227 0.013 0.006 0.008 876\n", " 228 0.031 0.012 0.018 967\n", " 229 0.034 0.015 0.021 815\n", " 230 0.030 0.011 0.016 796\n", " 231 0.037 0.014 0.020 725\n", " 232 0.046 0.015 0.023 738\n", " 233 0.038 0.012 0.018 673\n", " 234 0.011 0.003 0.005 646\n", " 235 0.018 0.005 0.008 627\n", " 236 0.036 0.009 0.014 561\n", " 237 0.017 0.004 0.007 492\n", " 238 0.006 0.002 0.003 493\n", " 239 0.026 0.007 0.011 449\n", " 240 0.050 0.011 0.018 458\n", " 241 0.030 0.008 0.012 395\n", " 242 0.045 0.010 0.017 388\n", " 243 0.030 0.006 0.010 338\n", " 244 0.000 0.000 0.000 317\n", " 245 0.073 0.013 0.022 303\n", " 246 0.000 0.000 0.000 276\n", " 247 0.000 0.000 0.000 255\n", " 248 0.059 0.012 0.020 249\n", " 249 0.021 0.004 0.006 274\n", " 250 0.000 0.000 0.000 239\n", " 251 0.056 0.010 0.017 200\n", " 252 0.000 0.000 0.000 260\n", " 253 0.043 0.014 0.021 293\n", " 254 0.011 0.003 0.004 398\n", " 255 0.327 0.320 0.324 2085\n", "\n", " accuracy 0.495 235500\n", " macro avg 0.020 0.015 0.016 235500\n", "weighted avg 0.423 0.495 0.454 235500\n", "\n" ] } ], "source": [ "# Print the precision and recall, among other metrics\n", "print(metrics.classification_report(testLabels.to_numpy().ravel(), \n", " y_pred2.ravel(), digits=3))" ] } ], "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": "3.11.5" } }, "nbformat": 4, "nbformat_minor": 5 }