数据挖掘十大算法(五)：SVM算法

    # encoding=utf-8
    
    import time
    
    import numpy as np
    import pandas as pd
    from sklearn.model_selection import train_test_split
    from sklearn.metrics import accuracy_score
    from sklearn import datasets
    from sklearn import svm
    
    if __name__ == '__main__':
    
    print('prepare datasets...')
    # Iris数据集
    # iris=datasets.load_iris()
    # features=iris.data
    # labels=iris.target
    
    # MINST数据集
    raw_data = pd.read_csv('../data/train_binary.csv', header=0)  # 读取csv数据，并将第一行视为表头，返回DataFrame类型
    data = raw_data.values
    
    features = data[0:10000:, 1::]
    labels = data[0:10000:, 0]    # 选取33%数据作为测试集，剩余为训练集
    print('原始数据feature的形状为：{}'.format(features.shape))
    print('原始数据label的形状为：{}'.format(labels.shape))
    
    train_features, test_features, train_labels, test_labels = train_test_split(features, labels, test_size=0.33, random_state=0)
    
    time_2 = time.time()
    print('Start training...')
    clf = svm.SVC()  # svm class   
    clf.fit(train_features, train_labels)  # training the svc model 
    time_3 = time.time()
    print('training cost %f seconds' % (time_3 - time_2))
    
    print('Start predicting...')
    test_predict = clf.predict(test_features)
    time_4 = time.time()
    print('predicting cost %f seconds' % (time_4 - time_3))
    
    score = accuracy_score(test_labels, test_predict)
    print("The accruacy score is %f" % score)

    import numpy as np
    import matplotlib.pyplot as plt
    from sklearn import svm
    
    # This is used for our dataset
    from sklearn.datasets import load_breast_cancer
    
    
    
    
    
    # 数据介绍：
    # The data[:, x:n] gets two features for the data given.
    # The : part gets all the rows in the matrix. And 0:2 gets the first 2 columns
    # If you want to get a different two features you can replace 0:2 with 1:3, 2:4,... 28:30,
    # there are 30 features in the set so it can only go up to 30.
    # If we wanted to plot a 3 dimensional plot then the difference between x and n needs to be 3 instead of two
    dataCancer = load_breast_cancer()
    data = dataCancer.data[:, 0:2]
    target = dataCancer.target
    
    
    
    # model
    model = svm.SVC(kernel='linear', C=1)
    model.fit(data, target)
    
    
    # plots the points
    plt.scatter(data[:, 0], data[:, 1], c=target, s=30, cmap=plt.cm.prism)
    
    
    # 下面是画决策边界图的常用画法
    # Creates the axis bounds for the grid
    axis = plt.gca()
    x_limit = axis.get_xlim()
    y_limit = axis.get_ylim()
    
    # Creates a grid to evaluate model
    x = np.linspace(x_limit[0], x_limit[1], 50)
    y = np.linspace(y_limit[0], y_limit[1], 50)
    X, Y = np.meshgrid(x, y)
    xy = np.c_[X.ravel(), Y.ravel()]
    
    
    
    # 这里可以得到边界，如果是要分类的话用model.predict
    decision_line = model.decision_function(xy).reshape(Y.shape)
    
    
    
    # Plot the decision line and the margins
    axis.contour(X, Y,  decision_line, colors='k',  levels=[0], linestyles=['-'])
    # Shows the support vectors that determine the desision line
    
    
    # 这个就是标记那些圆圈
    axis.scatter(model.support_vectors_[:, 0], model.support_vectors_[:, 1], s=100,
                             linewidth=1, facecolors='none', edgecolors='k')
    
    # Shows the graph
    plt.show()