A bug about python main_SemanticKITTI.py
ksama777 opened this issue · comments
I got a bug when I run python main_SemanticKITTI.py
**** EPOCH 000 ****
2022-08-01 17:14:40.759397
/media/wx/HDD/DQ/RandLA-Net-pytorch-master/RandLANet.py:265: UserWarning: torch.range is deprecated and will be removed in a future release because its behavior is inconsistent with Python's range builtin. Instead, use torch.arange, which produces values in [start, end).
reducing_list = torch.range(0, cfg.num_classes).long().cuda()
Traceback (most recent call last):
File "main_SemanticKITTI.py", line 205, in
train(start_epoch)
File "main_SemanticKITTI.py", line 185, in train
train_one_epoch()
File "main_SemanticKITTI.py", line 105, in train_one_epoch
loss, end_points = compute_loss(end_points, cfg)
File "/media/wx/HDD/DQ/RandLA-Net-pytorch-master/RandLANet.py", line 270, in compute_loss
loss = get_loss(valid_logits, valid_labels, cfg.class_weights)
File "/media/wx/HDD/DQ/RandLA-Net-pytorch-master/RandLANet.py", line 282, in get_loss
output_loss = criterion(logits, labels)
File "/home/wx/anaconda3/envs/RandLA-Net-pytorch-master/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/wx/anaconda3/envs/RandLA-Net-pytorch-master/lib/python3.7/site-packages/torch/nn/modules/loss.py", line 1152, in forward
label_smoothing=self.label_smoothing)
File "/home/wx/anaconda3/envs/RandLA-Net-pytorch-master/lib/python3.7/site-packages/torch/nn/functional.py", line 2846, in cross_entropy
return torch._C._nn.cross_entropy_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index, label_smoothing)
RuntimeError: weight tensor should be defined either for all 18 classes or no classes but got weight tensor of shape: [1, 19]
can you help me
thanks a lot
I have the same question with you.How can you save the question?
I have the same question with you.How can you save the question?
just change [1,19] to [19], and it work!
Hi ,
i helper_tool.py i changed this line
return np.expand_dims(ce_label_weight, axis=0) which is 2d array
to this one return ce_label_weight which is 1d array.
hope it helps you