使用python计算误差代码

import os import math ave_err=0.0 max_err=0.0 max_err_rate=0.0 length=0 with open("cpu_result.txt","r") as fp1, open("cuda_ppl_result.txt","r") as fp2: for l1 in fp1: l2 = fp2.readline() l1=l1[:-2] l2=l2[:-2] lst = l1.split(' ') lst2 = l2.split(' ') #print lst lst = [float(x) for x in lst ] length = length + len(lst) lst2 = [float(x) for x in lst2] #print (lst) #print (lst2)

        for index,x in enumerate(lst):
            y = lst2[index]
            ave_err = ave_err + abs(x-y)
            max_err = max(max_err,abs(x-y))
            max_err_rate = max(max_err_rate,abs(x-y)/x)

    print("len=",length)
    print("max_err=",max_err)
    print("max_err_rate=",max_err_rate*100,"%")
    print("ave_err=",ave_err/length)

需要提供两个文件，并且两个文件的数据格式相同。