txt保存的样本转为矩阵

1、读取指定文件内容

def file2matric(filename):
    """将txt文件的样本值存储到数组中 """
    with open(filename) as file_object:
        lines = file_object.readlines()#读取文件内容，返回存储每行的字符串列表
    lenOlines = len(lines) #文件行数
    line1 = lines[1].rstrip()
    listLine1 = line1.split('\t')
    columLine1 = len(listLine1)
    dataSetMat = zeros((lenOlines,columLine1-1))
    classLablesReturn = []
    index = 0
    for line in lines:
        line = line.rstrip()#去除每行末尾空格
        listFromLine = line.split('\t')
        dataSetMat[index,:] = listFromLine[0:-1]
        classLablesReturn.append(int(listFromLine[-1]))#标签应为整数类型
        index += 1
    return dataSetMat,classLablesReturn

2、读取指定文件夹下所有文件内容

from os import listdir

def imgtxts2Matric(filePath):
    imgFileList = listdir(filePath)#文件名列表，字符串类型
    filename0 = imgFileList[0]
    file0vector = imgtxt2Vector(filePath+"/"+filename0)
    m = len(imgFileList)#m个样本图片
    n = file0vector.shape[1]#矩阵列数，
    returnMat = zeros((m,n))
    returnLables = []
    for i in range(m):
        filename = imgFileList[i]
        imgVector = imgtxt2Vector(filePath+"/"+filename)
        returnMat[i,:] = imgVector[0,:]
        returnLables.append(int(filename[0]))
    return returnMat,returnLables