TypeError :当我在定义data_transformation对象的内部调用data_transformation函数时

Question

当我试图用我的get_data函数创建一个对象时：

   train = get_data(root ="My_train_path",
                     transform = data_transforms[TRAIN] )

它返回一个TypeError: 'function' object is not subscriptable。

data_dir = 'my_dataset_dir'

TEST  = 'test'
TRAIN = 'train'
VAL   = 'val'
def data_transforms(phase):
    if phase == TRAIN:
        transform = A.Compose([
            A.CLAHE(clip_limit=4.0, p=0.7),
            A.CoarseDropout(max_height=8, max_width=8, max_holes=8, p=0.5),
            A.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
            ToTensorV2(),
            
        ])
        
    if phase == VAL:
        transform = A.Compose([
            A.Resize(height=256,width=256),
            A.CenterCrop(height=224,width=224),
            A.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
            ToTensorV2(),
        ])
    
    if phase == TEST:
        transform = A.Compose([
            A.Resize(height=256,width=256),
            A.CenterCrop(height=224,width=224),
            A.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
            ToTensorV2(),
        ])        
        
    return transform

def get_data(root,transform):

  image_dataset = CustomImageFolder(root=".",
                                    transform = transform,
                                                         )
  return image_dataset





def make_loader(dataset, batch_size,shuffle,num_workers):
    loader = torch.utils.data.DataLoader(dataset=dataset,
                                         batch_size=batch_size, 
                                         shuffle=shuffle,
                                         pin_memory=True, num_workers=num_workers)
    return loader

Answer 1

问题就像错误消息说的那样：data_transforms是您定义的函数，您希望用训练阶段作为参数来调用它。但是，使用方括号([])错误地订阅了函数。要解决这个问题，请将方括号替换为括号(())，就像函数调用一样。

那是,

train = get_data(root ="My_train_path",
                 transform = data_transforms(TRAIN) )