diff --git a/Code/Day 11_K-NN.ipynb b/Code/Day 11_K-NN.ipynb new file mode 100644 index 0000000..8fdbe2d --- /dev/null +++ b/Code/Day 11_K-NN.ipynb @@ -0,0 +1,339 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#机器学习100天——第十一天：K近邻法（K-NN）" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第一步：导入相关库" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import pandas as pd" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第二步：导入数据集" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " User ID Gender Age EstimatedSalary Purchased\n", + "0 15624510 Male 19 19000 0\n", + "1 15810944 Male 35 20000 0\n", + "2 15668575 Female 26 43000 0\n", + "3 15603246 Female 27 57000 0\n", + "4 15804002 Male 19 76000 0\n", + "5 15728773 Male 27 58000 0\n", + "6 15598044 Female 27 84000 0\n", + "7 15694829 Female 32 150000 1\n", + "8 15600575 Male 25 33000 0\n", + "9 15727311 Female 35 65000 0\n", + "10 15570769 Female 26 80000 0\n", + "11 15606274 Female 26 52000 0\n", + "12 15746139 Male 20 86000 0\n", + "13 15704987 Male 32 18000 0\n", + "14 15628972 Male 18 82000 0\n", + "15 15697686 Male 29 80000 0\n", + "16 15733883 Male 47 25000 1\n", + "17 15617482 Male 45 26000 1\n", + "18 15704583 Male 46 28000 1\n", + "19 15621083 Female 48 29000 1\n", + "20 15649487 Male 45 22000 1\n", + "21 15736760 Female 47 49000 1\n", + "22 15714658 Male 48 41000 1\n", + "23 15599081 Female 45 22000 1\n", + "24 15705113 Male 46 23000 1\n", + "25 15631159 Male 47 20000 1\n", + "26 15792818 Male 49 28000 1\n", + "27 15633531 Female 47 30000 1\n", + "28 15744529 Male 29 43000 0\n", + "29 15669656 Male 31 18000 0\n", + ".. ... ... ... ... ...\n", + "370 15611430 Female 60 46000 1\n", + "371 15774744 Male 60 83000 1\n", + "372 15629885 Female 39 73000 0\n", + "373 15708791 Male 59 130000 1\n", + "374 15793890 Female 37 80000 0\n", + "375 15646091 Female 46 32000 1\n", + "376 15596984 Female 46 74000 0\n", + "377 15800215 Female 42 53000 0\n", + "378 15577806 Male 41 87000 1\n", + "379 15749381 Female 58 23000 1\n", + "380 15683758 Male 42 64000 0\n", + "381 15670615 Male 48 33000 1\n", + "382 15715622 Female 44 139000 1\n", + "383 15707634 Male 49 28000 1\n", + "384 15806901 Female 57 33000 1\n", + "385 15775335 Male 56 60000 1\n", + "386 15724150 Female 49 39000 1\n", + "387 15627220 Male 39 71000 0\n", + "388 15672330 Male 47 34000 1\n", + "389 15668521 Female 48 35000 1\n", + "390 15807837 Male 48 33000 1\n", + "391 15592570 Male 47 23000 1\n", + "392 15748589 Female 45 45000 1\n", + "393 15635893 Male 60 42000 1\n", + "394 15757632 Female 39 59000 0\n", + "395 15691863 Female 46 41000 1\n", + "396 15706071 Male 51 23000 1\n", + "397 15654296 Female 50 20000 1\n", + "398 15755018 Male 36 33000 0\n", + "399 15594041 Female 49 36000 1\n", + "\n", + "[400 rows x 5 columns]\n" + ] + } + ], + "source": [ + "dataset = pd.read_csv('../datasets/Social_Network_Ads.csv')\n", + "print(dataset)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "为了方便理解，这里我们只取Age年龄和EstimatedSalary估计工资作为特征" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [], + "source": [ + "X = dataset.iloc[:, [2, 3]].values\n", + "y = dataset.iloc[:, 4].values" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第三步：将数据划分成训练集和测试集" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.cross_validation import train_test_split\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 0)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第四步：特征缩放" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "D:\\anaconda\\lib\\site-packages\\sklearn\\utils\\validation.py:429: DataConversionWarning: Data with input dtype int64 was converted to float64 by StandardScaler.\n", + " warnings.warn(msg, _DataConversionWarning)\n" + ] + } + ], + "source": [ + "from sklearn.preprocessing import StandardScaler\n", + "sc = StandardScaler()\n", + "X_train = sc.fit_transform(X_train)\n", + "X_test = sc.transform(X_test)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "collapsed": true + }, + "source": [ + "##第五步：使用K-NN对训练集数据进行训练" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "从sklearn的neighbors类中导入KNeighborsClassifier学习器" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "from sklearn.neighbors import KNeighborsClassifier" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "设置好相关的参数\n", + "n_neighbors = 5(K值的选择，默认选择5)、\n", + "metric = 'minkowski'(距离度量的选择，这里选择的是闵氏距离(默认参数))、\n", + "p = 2 (距离度量metric的附属参数，只用于闵氏距离和带权重闵氏距离中p值的选择，p=1为曼哈顿距离， p=2为欧式距离。默认为2)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n", + " metric_params=None, n_jobs=1, n_neighbors=5, p=2,\n", + " weights='uniform')" + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "classifier = KNeighborsClassifier(n_neighbors = 5, metric = 'minkowski', p = 2)\n", + "classifier.fit(X_train, y_train)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第六步：对测试集进行预测" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 1 0 0 1 0 1 0 1 0 0 0 0 0 0 1 0 0 0 0\n", + " 0 0 1 0 0 0 0 1 0 0 1 0 1 1 0 0 1 1 1 0 0 1 0 0 1 0 1 0 1 0 0 0 0 1 0 0 1\n", + " 0 0 0 0 1 1 1 1 0 0 1 0 0 1 1 0 0 1 0 0 0 0 0 1 1 1]\n" + ] + } + ], + "source": [ + "y_pred = classifier.predict(X_test)\n", + "print(y_pred)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "##第七步：生成混淆矩阵" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "混淆矩阵可以对一个分类器性能进行分析，由此可以计算出许多指标，例如：ROC曲线、正确率等" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[[64 4]\n", + " [ 3 29]]\n" + ] + } + ], + "source": [ + "from sklearn.metrics import confusion_matrix\n", + "cm = confusion_matrix(y_test, y_pred)\n", + "print(cm)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + " 预测值\n", + " 0 1\n", + " 实0 64 4 \n", + " 际1 3 29\n", + " 值\n", + "\n", + "预测集中的0总共有68个，1总共有32个。\n", + "在这个混淆矩阵中，实际有68个0，但K-NN预测出有67(64+3)个0，其中有3个实际上是1。\n", + "同时K-NN预测出有33(4+29)个1，其中4个实际上是0。" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.6.1" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}