{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "aaea6e74",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Yang Lu\n",
    "# import Keras & Tensorflow\n",
    "import tensorflow as tf\n",
    "import pandas as pd\n",
    "from tensorflow import keras\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c9d2752f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def TrainKNN():\n",
    "\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "    # 加载数据集 Load Dataset\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "    # Load image data 下载数据集\n",
    "    fashion_mnist = keras.datasets.fashion_mnist\n",
    "    # 载入训练集和测试集（默认顺序）\n",
    "    (X_train, Y_train), (X_test, Y_test) = fashion_mnist.load_data()\n",
    "    # 定义名称list（方便label转名称）\n",
    "    # 即：在代码上补充一下每个类别的名称,组成一个list,代码如下:\n",
    "    class_names = [\"T-shirt\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\", \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]\n",
    "    # 打印图片\n",
    "   #显示多张照片\n",
    "    plt.figure(figsize=(10,10))\n",
    "    for i in range(25):\n",
    "        plt.subplot(5,5,i+1)\n",
    "        plt.xticks([])\n",
    "        plt.yticks([])\n",
    "        plt.grid(False)\n",
    "        plt.imshow(X_train[i], cmap=plt.cm.binary)\n",
    "        plt.xlabel(class_names[Y_train[i]])\n",
    "\n",
    "\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "    # 处理数据部分 Processing data section\n",
    "    # X_train = X_train.astype('float32') --- 将数据类型转为32位浮点数 Convert data type to 32-bit floating point number\n",
    "    # X_train /= 255. --- 归一化处理，将原来数据（0-255）---> (0-1) --- 减小计算量\n",
    "    # Normalization processing: reduce the calculation amount of the original data (0-255) --->(0-1)\n",
    "    # X_train = X_train.reshape(60000, 784) --- 改变训练数据的形状 --- 图片的大小是28x28，60000张训练\n",
    "    # Change the shape of training data --- the size of the image is 28x28   60000 pieces of training\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "    X_train = X_train.astype('float32')\n",
    "    X_test = X_test.astype('float32')\n",
    "\n",
    "    X_train /= 255.\n",
    "    X_test /= 255.\n",
    "\n",
    "    X_train = X_train.reshape(60000, 784)\n",
    "    X_test = X_test.reshape(10000, 784)\n",
    "\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "    # 训练数据部分\n",
    "    # knn = KNeighborsClassifier(n_neighbors=10) --- 初始化knn Initialize knn\n",
    "    # knn.fit(X_train, Y_train) --- 将训练数据喂入knn，训练里面的参数，达到最适\n",
    "    # Feed the training data into knn, and train the parameters to reach the optimum\n",
    "    # knn.predict(X_test) --- 使用训练好的knn，测试数据集\n",
    "    # Use trained knn and test data set\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "    knn = KNeighborsClassifier(n_neighbors=10)\n",
    "    knn.fit(X_train, Y_train)\n",
    "    Y_pred = knn.predict(X_test)\n",
    "\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "    # label_names --- 类别名称\n",
    "    # classification_report_result --- 得到结果报告\n",
    "    # ------------------------------------------------------------------------------------------------------------------\n",
    "    C2= confusion_matrix(Y_test, Y_pred)\n",
    "    print(\"Confusion_matrix of KNN:\")\n",
    "    print(C2)\n",
    "    print(\"-------------------------------------------------------------------\")\n",
    "    print(\"Other evaluation indicators of KNN:\")\n",
    "    classification_report_result = classification_report(Y_test, Y_pred, target_names=class_names, digits=4)\n",
    "    print(classification_report_result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1a39faa4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion_matrix of KNN:\n",
      "[[863   0  20  19   5   0  84   1   8   0]\n",
      " [  8 964   6  13   4   0   4   0   1   0]\n",
      " [ 21   2 813   7  82   0  74   0   1   0]\n",
      " [ 37   4  14 871  37   0  35   0   2   0]\n",
      " [  1   0 128  25 755   0  90   0   1   0]\n",
      " [  1   0   0   1   0 807   5 100   2  84]\n",
      " [195   0 135  22  76   0 560   0  12   0]\n",
      " [  0   0   0   0   0   2   0 969   0  29]\n",
      " [  1   1  14   3   8   0  16   6 949   2]\n",
      " [  0   0   0   0   0   1   1  34   0 964]]\n",
      "-------------------------------------------------------------------\n",
      "Other evaluation indicators of KNN:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "     T-shirt     0.7657    0.8630    0.8115      1000\n",
      "     Trouser     0.9928    0.9640    0.9782      1000\n",
      "    Pullover     0.7195    0.8130    0.7634      1000\n",
      "       Dress     0.9063    0.8710    0.8883      1000\n",
      "        Coat     0.7808    0.7550    0.7677      1000\n",
      "      Sandal     0.9963    0.8070    0.8917      1000\n",
      "       Shirt     0.6444    0.5600    0.5993      1000\n",
      "     Sneaker     0.8730    0.9690    0.9185      1000\n",
      "         Bag     0.9723    0.9490    0.9605      1000\n",
      "  Ankle boot     0.8934    0.9640    0.9274      1000\n",
      "\n",
      "    accuracy                         0.8515     10000\n",
      "   macro avg     0.8545    0.8515    0.8506     10000\n",
      "weighted avg     0.8545    0.8515    0.8506     10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -----------------------------------------------------------------------------------------------------------------------\n",
    "# 主函数\n",
    "# -----------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "if __name__ == '__main__':\n",
    "    knn = TrainKNN()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "614ea871",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}