数据从txt文本导入python

机器学习实战 p21

源代码：

def file2matrix(filename):
    fr = open(filename)
    numberOfLines = len(fr.readlines())         #get the number of lines in the file
    returnMat = zeros((numberOfLines,3))        #prepare matrix to return
    classLabelVector = []                       #prepare labels return   
    fr = open(filename)
    index = 0
    for line in fr.readlines():
        line = line.strip()
        listFromLine = line.split('\t')
        returnMat[index,:] = listFromLine[0:3]
        classLabelVector.append(int(listFromLine[-1]))  此句报错
        index += 1
    return returnMat,classLabelVector

报错如下：

>>> mat,label = kNN.file2matrix('datingTestSet.txt')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "kNN.py", line 50, in file2matrix
    classLabelVector.append(int(listFromLine[-1]))
ValueError: could not convert string to int: largeDoses

解决方法：

listFromLine[-1]的值形似如下格式，带有回车换行符

largeDoses\r\n

smallDoses\r\n

didntLike\r\n

didntLike\r\n

didntLike\r\n

要将字母字符串转换为int类型是不可能的。

作者定义largeDoses 为3，smallDoses 为2，didntLike为1

于是笔者增加了一个字典类型

d = {'didntLike': 1, 'smallDoses': 2, 'largeDoses': 3}

通过d[listFromLine[-1]]得到对应的label

更改后的代码如下：

rf.py

from numpy import *
import operator
from os import listdir

def rf(filename):
    fr = open(filename)
    numberOfLines = len(fr.readlines())         #get the number of lines in the file
    returnMat = zeros((numberOfLines,3))        #prepare matrix to return
    d = {'didntLike': 1, 'smallDoses': 2, 'largeDoses': 3}
    classLabelVector = []
    index = 0
    fr = open(filename)
    for line in fr.readlines():
        listFromLine = line.split('\t')
        returnMat[index,:] = listFromLine[0:3]
        listFromLine[-1] = listFromLine[-1][0:-2]        #去除尾端的回车换行符
        classLabelVector.append(d[listFromLine[-1]])   #取到字典中对应的label值
        index += 1
    return returnMat,classLabelVector

画图：

import rf
mat,label = rf.rf('datingTestSet.txt')
import matplotlib
import matplotlib.pyplot as plt
fig = plt.figure()


>>> ax1 = fig.add_subplot(2, 2, 1)
>>> ax1.scatter(mat[:,0],mat[:,1])
>>> ax2 = fig.add_subplot(2, 2, 2)
>>> ax2.scatter(mat[:,1],mat[:,2])

from numpy import array        #需要自己导入array，否则会报错
>>> ax3 = fig.add_subplot(2, 2, 3)
>>> ax3.scatter(mat[:,0],mat[:,1],15.0*array(label),15.0*array(label))
ax4 = fig.add_subplot(2, 2, 4)
ax4.scatter(mat[:,1],mat[:,2],15.0*array(label),15.0*array(label))
plt.show()