机器学习 machine learning

1.项目介绍

机器学习（Machine Learning）正在广泛应用于各种领域，不断学习，充实自己，才能跟上步伐。

在Gihub发现了一个项目，作者写了自己的学习过程，图文并茂，由浅入深，有完整的Python代码，非常值得借鉴。但英文读起来不方便，因此，我和小伙伴一起，把它翻译成了中文，截至目前，包含以下内容：

数据预处理
简单线性回归
多元线性回归
逻辑回归
k近邻法（k-NN）
支持向量机（SVM）
决策树
随机森林

项目地址：

https://github.com/MachineLearning100/100-Days-Of-ML-Code

2.例子

支持向量机（support vector machine）：

Python代码：

  1 #Day13: Support Vector Machine (SVM)
  2 
  3  
  4 
  5 #Importing the libraries
  6 
  7 import numpy as np
  8 
  9 import matplotlib.pyplot as plt
 10 
 11 import pandas as pd
 12 
 13  
 14 
 15 #Importing the dataset
 16 
 17 dataset = pd.read_csv('../datasets/Social_Network_Ads.csv')
 18 
 19 X = dataset.iloc[:, [2, 3]].values
 20 
 21 y = dataset.iloc[:, 4].values
 22 
 23  
 24 
 25 #Splitting the dataset into the Training set and Test set
 26 
 27 from sklearn.cross_validation import train_test_split
 28 
 29 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 0)
 30 
 31  
 32 
 33 #Feature Scaling
 34 
 35 from sklearn.preprocessing import StandardScaler
 36 
 37 sc = StandardScaler()
 38 
 39 X_train = sc.fit_transform(X_train)
 40 
 41 X_test = sc.transform(X_test)
 42 
 43  
 44 
 45 #Fitting SVM to the Training set
 46 
 47 from sklearn.svm import SVC
 48 
 49 classifier = SVC(kernel = 'linear', random_state = 0)
 50 
 51 classifier.fit(X_train, y_train)
 52 
 53  
 54 
 55 #Predicting the Test set results
 56 
 57 y_pred = classifier.predict(X_test)
 58 
 59  
 60 
 61 #Making the Confusion Matrix
 62 
 63 from sklearn.metrics import confusion_matrix
 64 
 65 cm = confusion_matrix(y_test, y_pred)
 66 
 67  
 68 
 69 #Visualising the Training set results
 70 
 71 from matplotlib.colors import ListedColormap
 72 
 73 X_set, y_set = X_train, y_train
 74 
 75 X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:, 0].max() + 1, step = 0.01),
 76 
 77                      np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:, 1].max() + 1, step = 0.01))
 78 
 79 plt.contourf(X1, X2, classifier.predict(np.array([X1.ravel(), X2.ravel()]).T).reshape(X1.shape),
 80 
 81              alpha = 0.75, cmap = ListedColormap(('red', 'green')))
 82 
 83 plt.xlim(X1.min(), X1.max())
 84 
 85 plt.ylim(X2.min(), X2.max())
 86 
 87 for i, j in enumerate(np.unique(y_set)):
 88 
 89     plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
 90 
 91                 c = ListedColormap(('red', 'green'))(i), label = j)
 92 
 93 plt.title('SVM (Training set)')
 94 
 95 plt.xlabel('Age')
 96 
 97 plt.ylabel('Estimated Salary')
 98 
 99 plt.legend()
100 
101 plt.show()
102 
103  
104 
105 #Visualising the Test set results
106 
107 from matplotlib.colors import ListedColormap
108 
109 X_set, y_set = X_test, y_test
110 
111 X1, X2 = np.meshgrid(np.arange(start = X_set[:, 0].min() - 1, stop = X_set[:, 0].max() + 1, step = 0.01),
112 
113                      np.arange(start = X_set[:, 1].min() - 1, stop = X_set[:, 1].max() + 1, step = 0.01))
114 
115 plt.contourf(X1, X2, classifier.predict(np.array([X1.ravel(), X2.ravel()]).T).reshape(X1.shape),
116 
117              alpha = 0.75, cmap = ListedColormap(('red', 'green')))
118 
119 plt.xlim(X1.min(), X1.max())
120 
121 plt.ylim(X2.min(), X2.max())
122 
123 for i, j in enumerate(np.unique(y_set)):
124 
125     plt.scatter(X_set[y_set == j, 0], X_set[y_set == j, 1],
126 
127                 c = ListedColormap(('red', 'green'))(i), label = j)
128 
129 plt.title('SVM (Test set)')
130 
131 plt.xlabel('Age')
132 
133 plt.ylabel('Estimated Salary')
134 
135 plt.legend()
136 
137 plt.show()

输出结果：