文件输入（fileinput.input()）函数给出意外输出

Question

我使用这个脚本删除重复的行和大于10000的值。你知道吗

import fileinput
import os, sys
import re

def get_immediate_subdirectories(a_dir):
   return [name for name in os.listdir(a_dir) if os.path.isdir(os.path.join(a_dir,name))]

for i in get_immediate_subdirectories(os.getcwd()+'\population_vs_time\\'):
  for j in get_immediate_subdirectories(os.getcwd()+'\population_vs_time\\'+i):        
    for file in os.listdir(os.getcwd()+'\population_vs_time\\'+i+'\\'+j):
        seen=set()
        my_dir=os.getcwd()+'\population_vs_time\\'+str(i)+'\\'+str(j)+'\\'+file
        matches=re.match("time",str(file))
        if not matches:
            print (my_dir)
            f = fileinput.input(files=my_dir)

            for line in f:
                if line in seen: continue  # skip duplicate
                flag=0
                words = line.split()
                for word in words:
                    # try:
                    i=float(word)
                    if i>10000:
                        flag=1
                        break
                    # except ValueError:
                    #     flag=1
                if flag==1: continue
                seen.add(line)
                print line, # standard output is now redirected to the file

            f.close()

我有一个类型为string的\dir变量，使用print函数来显示这个变量的值，它为前三个文件提供了正确的输出

C:\Program Files (x86)\Guimoo\bin\population_vs_time\mocmaes\frontsize_gen=220\mocmaes_gen_220_100_.csv
C:\Program Files (x86)\Guimoo\bin\population_vs_time\mocmaes\frontsize_gen=220\mocmaes_gen_220_120_.csv
C:\Program Files (x86)\Guimoo\bin\population_vs_time\mocmaes\frontsize_gen=220\mocmaes_gen_220_140_.csv

但当读取下一个文件时，它会给出随机数作为输出（1.70512而不是mocmaes）

C:\Program Files (x86)\Guimoo\bin\population_vs_time\1.70512\frontsize_gen=220\mocmaes_gen_220_160_.csv

我想我缺少python\escape角色的基本知识。是这样吗？你知道吗

注：mocmaes文件夹的截图附在