SVM对sklearn自带手写数字数据集进行分类

sklearn自带一些数据集，其中手写数字数据集可通过load_digits加载，我找到load_digits里头是这样

def load_linnerud():    """Load and return the linnerud dataset (multivariate regression).    Samples total: 20    Dimensionality: 3 for both data and targets    Features: integer    Targets: integer    Returns    -------    data : Bunch        Dictionary-like object, the interesting attributes are: 'data' and        'targets', the two multivariate datasets, with 'data' corresponding to        the exercise and 'targets' corresponding to the physiological        measurements, as well as 'feature_names' and 'target_names'.    """    base_dir = join(dirname(__file__), 'data/')    # Read data    data_exercise = np.loadtxt(base_dir + 'linnerud_exercise.csv', skiprows=1)    data_physiological = np.loadtxt(base_dir + 'linnerud_physiological.csv',                                    skiprows=1)    # Read header    with open(base_dir + 'linnerud_exercise.csv') as f:        header_exercise = f.readline().split()    with open(base_dir + 'linnerud_physiological.csv') as f:        header_physiological = f.readline().split()    with open(dirname(__file__) + '/descr/linnerud.rst') as f:        descr = f.read()    return Bunch(data=data_exercise, feature_names=header_exercise,                 target=data_physiological,                 target_names=header_physiological,                 DESCR=descr)

有数据说明，很详细，也可以看到实际那批数据来自一个csv文件，可以按照提示挑一个样例show一下看看

import pylab as pldigits = load_digits()# 数据纬度，1797幅图，8*8# 显示一副图片# pl.gray()# pl.matshow(digits.images[0])# pl.show()

言归正传，调用sklearn的svm包里的LinearSVC做一下分类实验

# -*- coding:utf-8 -*-import sysfrom sklearn.datasets import load_digits  # 加载手写数字识别数据import pylab as plfrom sklearn.cross_validation import train_test_split  # 训练测试数据分割from sklearn.preprocessing import StandardScaler  # 标准化工具from sklearn.svm import LinearSVCfrom sklearn.metrics import classification_report  # 预测结果分析工具reload(sys)sys.setdefaultencoding('utf-8')digits = load_digits()# 数据纬度，1797幅图，8*8print digits.data.shape# 分割数据X_train, X_test, Y_train, Y_test = train_test_split(digits.data, digits.target, test_size=0.25, random_state=33)ss = StandardScaler()# fit是实例方法，必须由实例调用X_train = ss.fit_transform(X_train)X_test = ss.transform(X_test)lsvc = LinearSVC()lsvc.fit(X_train, Y_train)Y_predict = lsvc.predict(X_test)print classification_report(Y_test, Y_predict, target_names=digits.target_names.astype(str))

训练集与测试集1比3分割，分类预测的性能如下：

             precision    recall  f1-score   support          0       0.92      1.00      0.96        35          1       0.96      0.98      0.97        54          2       0.98      1.00      0.99        44          3       0.93      0.93      0.93        46          4       0.97      1.00      0.99        35          5       0.94      0.94      0.94        48          6       0.96      0.98      0.97        51          7       0.92      1.00      0.96        35          8       0.98      0.84      0.91        58          9       0.95      0.91      0.93        44avg / total       0.95      0.95      0.95       450

需要记住几个性能指标的含义

precision为精确率，是真阳结果与所有被判为阳性结果的比。

recall为召回率， 是真阳结果与所有实际为阳性数据的比。

真阳者，实际为真且判为真。