pytorch加载文件夹顺序

• 文件夹顺序

from torchvision import datasets

data_dir=$DATA
train_datasets = datasets.ImageFolder(os.path.join(data_dir, 'train')
print(train_datasets.classes, train_datasets.class_to_idx)

['negative', 'others', 'positive'],{'negative': 0, 'others': 1, 'positive': 2}

• 内部读取逻辑

def find_classes(dir):
    classes = [d for d in os.listdir(dir) if os.path.isdir(os.path.join(dir, d))]
    classes.sort()
    class_to_idx = {classes[i]: i for i in range(len(classes))}
    return classes, class_to_idx

• 直接获取文件夹顺序

ls $DATA

negative others positive

 

参考文献：

1. Torchvision.datasets中的ImageFolder函数详解