python面向对象编程

用面向对象的方法进行数据的处理

昨天学习了面向对象的基础，尝试去用该思想去预处理机器学习的数据。基本思路是往函数中输入数据文件名称和分配的比例ratio，输出测试集，测试标签，训练集，训练标签。

from numpy import *  
from sklearn import svm    #库不能写在函数中。
class FunSVm():   #没参数
    def __init__(self,name,ratio):  #定义需要给funsvm传入2个参数，名字和比例
        self.name=name
        self.ratio=ratio   #定义类变量
        
    def file2mat(self): 
        with open(self.name) as fr:  
            arrayLine = fr.readlines()   #读取文件
        linenum = len(arrayLine)
        dataMat = zeros((linenum,8))
        label = []
        index = 0
        for line in arrayLine:
            line = line.strip()
            line = line.split(",")
            dataMat[index,:] = line[0:8]
            label.append(int(line[-1]))
            index += 1
        return dataMat,label,linenum
            
    def classify(self):
        ratio = self.ratio
        x,y,length = self.file2mat()
        trainNum = int(ratio * length) 
        x_train = x[:trainNum]
        x_test = x[trainNum:]
        y_train = y[:trainNum]
        y_test = y[trainNum:]
        return x_train,x_test,y_train,y_test

data = FunSVm("pima-indians-diabetes.txt",0.8)

array([[   6.   ,  148.   ,   72.   , ...,   33.6  ,    0.627,   50.   ],    #输出numpy矩阵
       [   1.   ,   85.   ,   66.   , ...,   26.6  ,    0.351,   31.   ],
       [   8.   ,  183.   ,   64.   , ...,   23.3  ,    0.672,   32.   ],
       ..., 
       [   3.   ,  174.   ,   58.   , ...,   32.9  ,    0.593,   36.   ],
       [   7.   ,  168.   ,   88.   , ...,   38.2  ,    0.787,   40.   ],
       [   6.   ,  105.   ,   80.   , ...,   32.5  ,    0.878,   26.   ]])