RelTR icon indicating copy to clipboard operation
RelTR copied to clipboard

Cannot run with pytorch 1.13.1 and cuda 11.7

Open IceIce1ce opened this issue 4 months ago • 0 comments

Hi author,

When I run with the above environment, an error happend:

terminate called after throwing an instance of 'std::runtime_error' what(): NCCL Error 1: unhandled cuda error Traceback (most recent call last): File "main.py", line 239, in main(args) File "main.py", line 207, in main test_stats, coco_evaluator = evaluate(model, criterion, postprocessors, data_loader_val, base_ds, device, args) File "/home/vsw/miniconda3/envs/detr/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/home/vsw/Desktop/RelTR/engine.py", line 114, in evaluate loss_dict = criterion(outputs, targets) File "/home/vsw/miniconda3/envs/detr/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(*input, **kwargs) File "/home/vsw/Desktop/RelTR/models/reltr.py", line 298, in forward indices = self.matcher(outputs_without_aux, targets) File "/home/vsw/miniconda3/envs/detr/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl return forward_call(*input, **kwargs) File "/home/vsw/miniconda3/envs/detr/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/home/vsw/Desktop/RelTR/models/matcher.py", line 92, in forward indices = [linear_sum_assignment(c[i]) for i, c in enumerate(C.split(sizes, -1))] File "/home/vsw/Desktop/RelTR/models/matcher.py", line 92, in indices = [linear_sum_assignment(c[i]) for i, c in enumerate(C.split(sizes, -1))] ValueError: matrix contains invalid numeric entries

IceIce1ce avatar Aug 09 '25 09:08 IceIce1ce