python学习

“人生苦短，我用python”

这几日折腾搭建了python平台，主要安装了numpy，scipy，matplotlib，scikit-learn这几个包。

先来一个小程序练练手。

from numpy import *
import sklearn
import random
from sklearn import linear_model

# load dataset
infile = "C:\\Users\\Administrator\\Desktop\\1.txt"
fr = open(infile)
lines = fr.readlines()
numberOfLines = len(lines)
src = []
result = []
for line in lines:
    listFromLine = line.strip().split('\t')
    # map all elements to float
    floatLine = map(float, listFromLine)
    src.append(floatLine)
    
src = array(src)
row, col = src.shape
for loop in range(0, 10):
    
    # shuffle the sample
    random.shuffle(src)
    
    trainFeature = src[0 : row / 2, 0 : -1]
    trainLabel = src[0 : row / 2, -1]
    testFeature = src[row / 2 : row, 0 : -1]
    testLabel = src[row / 2 : row, -1]
    testNumber = len(testLabel)
    
    # train
    clf = linear_model.LogisticRegression()
    clf.fit(trainFeature, trainLabel)
    
    # test
    Y = clf.predict(testFeature)

    # predict answer
    right = 0
    for i in range(0, testNumber):
        if Y[i] == testLabel[i]:
            right = right + 1
    result.append(float(right) / float(testNumber) * 100)
    
print sum(result) / len(result)