Problem in training

[Johumliu]

[ep 1][it 80][loss 0.0043][lr 1.0000][0.71s]
[cnt: gt: 155.0 pred: 0.03]
[ep 1][it 90][loss 0.0047][lr 1.0000][0.72s]
[cnt: gt: 10.0 pred: 0.29]
Traceback (most recent call last):
File "train.py", line 33, in
cc_trainer.forward()
File "/root/ljh/NWPU-Crowd-Sample-Code/trainer.py", line 63, in forward
self.train()
File "/root/ljh/NWPU-Crowd-Sample-Code/trainer.py", line 82, in train
for i, data in enumerate(self.train_loader, 0):
File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py", line 310, in next
data = self.dataset_fetcher.fetch(index) # may raise StopIteration
File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/root/ljh/NWPU-Crowd-Sample-Code/datasets/basedataset.py", line 57, in getitem
img = self.img_transform(img)
File "/usr/local/lib/python3.6/dist-packages/torchvision/transforms/transforms.py", line 61, in call
img = t(img)
File "/usr/local/lib/python3.6/dist-packages/torchvision/transforms/transforms.py", line 166, in call
return F.normalize(tensor, self.mean, self.std, self.inplace)
File "/usr/local/lib/python3.6/dist-packages/torchvision/transforms/functional.py", line 217, in normalize
tensor.sub(mean[:, None, None]).div(std[:, None, None])
RuntimeError: The size of tensor a (4) must match the size of tensor b (3) at non-singleton dimension 0

[gjy3035]

Is this error re-produced?