ValueError: operands could not be broadcast together with shapes (900,1600,1) (900,1600)

[largerwxt]

Hello, the following error occurred when I reproduced the code. How can I solve it?
Traceback (most recent call last):
File "core/train.py", line 156, in
main()
File "core/train.py", line 150, in main
train(args, model, optimizer, train_loader, epoch)
File "core/train.py", line 78, in train
for iteration, batch in enumerate(train_loader, 1):
File "/home/anaconda/envs/wxt-env/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 628, in next
data = self._next_data()
File "/home/anaconda/envs/wxt-env/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 671, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/home/anaconda/envs/wxt-env/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 58, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/anaconda/envs/wxt-env/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 58, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/wxt/AIM-master/core/data.py", line 101, in getitem
ori, fg, bg = generate_composite(fg, bg, mask)
File "/home/wxt/AIM-master/core/util.py", line 111, in generate_composite
composite = alpha * fg + (1 - alpha) * bg
ValueError: operands could not be broadcast together with shapes (900,1600,1) (900,1600)