When training the network, there is a strange error.

[GuangwuOfLongCity]

When I run python train.py, there is a strange error.

Epoch 0: StepDecay set learning rate to 0.05000000074505806.
Traceback (most recent call last):
  File "train.py", line 212, in <module>
    train()
  File "train.py", line 145, in train
    for step, (lr_patch, hr_patch) in enumerate(train_ds):
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\dataflow\utils.py", line 417, in __next__
    data = self._next_data()
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\dataflow\utils.py", line 438, in _next_data
    data = self._dataset_fetcher.fetch(index)
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\dataflow\utils.py", line 347, in fetch
    data = [self.dataset[id] for id in batch_indices]
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\dataflow\utils.py", line 347, in <listcomp>
    data = [self.dataset[id] for id in batch_indices]
  File "train.py", line 48, in __getitem__
    hr_patch = self.hr_trans(img)
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\vision\transforms\transforms.py", line 154, in __call__
    data = t(data)
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\vision\transforms\transforms.py", line 850, in __call__
    return random_crop(
  File "C:\ProgramData\Anaconda3\envs\SRGAN\lib\site-packages\tensorlayerx\vision\transforms\functional.py", line 760, in random_crop
    offset_height = np.random.randint(0, h - target_height)
  File "mtrand.pyx", line 746, in numpy.random.mtrand.RandomState.randint
  File "_bounded_integers.pyx", line 1338, in numpy.random._bounded_integers._rand_int32
ValueError: high <= 0

Could you help me?
Thanks, very much!
@zsdonghao