用python制作训练集和测试集的图片名列表文本

# -*- coding: utf-8 -*-
from pathlib import Path  #从pathlib中导入Path
import os
import fileinput
import random
root_path='/home/tay/Videos/trash/垃圾分类项目/total/'
train =  open('./trash_train.txt','a')
test = open('./trash_test.txt','a')
pwd = os.getcwd() +'/'# the val data path 训练集的路径


def gen_txt():
    i =0
    for file in os.listdir(root_path):
        print('file is{}'.format(str(file)))
        for init in os.listdir(os.path.join(root_path, file)): #子文件夹
            print('init is{}'.format(str(init)))
            i += 1
            pathDir = os.listdir(os.path.join(root_path, file, init)) #
            print('pathDir is', pathDir)
            file_num = len(pathDir)
            rate = 0.2
            pick_num = int(file_num * rate)
            sample = random.sample(pathDir, pick_num)  #随机选取20%的pathDir字符串
            print('sample is', sample)
            for pick_name in sample:
                test.write(root_path.split('total/')[-1] +file + '/' + init +'/' + pick_name + ' ' + str(i) + '
')
            # for name in pathDir: #文件夹中的图片名
                # print('name is{}'.format(str(name)))
                # if test
                # total.write(root_path.split('total/')[-1] +file + '/' + init +'/' + name + ' ' + str(i) + '
' )
            same = [x for x in pathDir if x in sample] #列表中相同的内容
            diff = [y for y in (sample + pathDir) if y not in same] #列表中不同的内容
            print('different', diff)
            print('same', same)
            for train_name in diff:
                train.write(root_path.split('total/')[-1] +file + '/' + init +'/' + train_name + ' ' + str(i) + '
')
gen_txt()

采用了random.sample函数来随机选取特定数量的文件名作为测试集，通过比较两个列表中不同的元素来获取训练集的文件名。

总体上就是在进行字符串操作。