import os
import random

'''
对图片数据集进行随机分类
以8: 1: 1的比例分为训练数据集，验证数据集和测试数据集
运行后在ImageSets文件夹中会出现四个文件
'''
ROOT = '/home/hhit/source/datasets/anquanmao/'
trainval_percent = 0.9
train_percent = 0.9
xmlfilepath = ROOT + 'Annotations'

#创建 ImageSets/Main 目录
imagesets = ROOT + 'ImageSets'
imagesets_main = ROOT + 'ImageSets/Main'
if not os.path.exists(imagesets):
	os.mkdir(imagesets)
if not os.path.exists(imagesets_main):
	os.mkdir(imagesets_main)
     
# 获取该路径下所有文件的名称，存放在list中
total_xml = os.listdir(xmlfilepath)

num = len(total_xml)
list = range(num)
tv = int(num * trainval_percent)
tr = int(tv * train_percent)
trainval = random.sample(list, tv)
train = random.sample(trainval, tr)

ftrainval = open(imagesets_main + '/trainval.txt', 'w')
ftest = open(imagesets_main + '/test.txt', 'w')
ftrain = open(imagesets_main + '/train.txt', 'w')
fval = open(imagesets_main + '/val.txt', 'w')

for i in list:
    # 获取文件名称中.xml之前的序号
    name = total_xml[i][:-4] + '\n'
    if i in trainval:
        ftrainval.write(name)
        if i in train:
            ftrain.write(name)
        else:
            fval.write(name)
    else:
        ftest.write(name)

ftrainval.close()
ftrain.close()
fval.close()
ftest.close()