import os
import random
filepath=r'C:\models\research\deeplab\datasets\Headshoulder_dataset\dataset\JPEGImages'
saveBasePath=r"C:\models\research\deeplab\datasets\Headshoulder_dataset\dataset"
trainval_percent=1
train_percent=0.7
total_xml = os.listdir(filepath)
#a=total_xml[1].split('.jpg')
#print(a[0])
num=len(total_xml)
print(num)
#assert 0
list=range(num)
tv=int(num*trainval_percent)
tr=int(tv*train_percent)
trainval= random.sample(list,tv)
train=random.sample(trainval,tr)
print("train and val size:",tv)
print("train size:",tr)
ftrainval = open(os.path.join(saveBasePath,'ImageSets/trainval.txt'), 'w')
ftrain = open(os.path.join(saveBasePath,'ImageSets/train.txt'), 'w')
fval = open(os.path.join(saveBasePath,'ImageSets/val.txt'), 'w')
for i in list:
a=total_xml[i].split('.jpg')
name=a[0]+'\n'
if i in trainval:
ftrainval.write(name)
if i in train:
ftrain.write(name)
else:
fval.write(name)
ftrainval.close()
ftrain.close()
fval.close()
制作PASCAL VOC格式的分割数据集,生成trainval.txt, train.txt, val.txt文件
最新推荐文章于 2023-08-07 13:30:14 发布