PyCenterNet icon indicating copy to clipboard operation
PyCenterNet copied to clipboard

Error in validation

Open YilinGao-SHU opened this issue 2 years ago • 1 comments

I'm sorry to bother you again, following your advice I managed to complete the training section, but when I did the validation, a new error was reported.

File "/server8/gyl/Data_clean/PyCenterNet-master/code/mmdet/models/detectors/base.py", line 183, in forward return self.forward_test(img, img_metas, **kwargs) File "/server8/gyl/Data_clean/PyCenterNet-master/code/mmdet/models/detectors/base.py", line 160, in forward_test return self.simple_test(imgs[0], img_metas[0], **kwargs) File "/server8/gyl/Data_clean/PyCenterNet-master/code/mmdet/models/detectors/single_stage.py", line 120, in simple_test outs, img_metas, rescale=rescale) File "/server8/gyl/Data_clean/PyCenterNet-master/code/mmdet/models/dense_heads/pycenternet_head.py", line 1082, in get_bboxes nms) File "/server8/gyl/Data_clean/PyCenterNet-master/code/mmdet/models/dense_heads/pycenternet_head.py", line 1204, in _get_bboxes_single tl_bboxes = torch.stack([x1, y1, x2, y2], dim=-1) RuntimeError: cuda runtime error (700) : an illegal memory access was encountered at /pytorch/aten/src/THC/THCCachingHostAllocator.cpp:278 terminate called after throwing an instance of 'c10::Error' what(): CUDA error: an illegal memory access was encountered (insert_events at /pytorch/c10/cuda/CUDACachingAllocator.cpp:771) frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x46 (0x7f37635b3536 in /nvme1/anaconda3/envs/gyl_mmdetection/lib/python3.7/site-packages/torch/lib/libc10.so) frame #1: c10::cuda::CUDACachingAllocator::raw_delete(void) + 0x7ae (0x7f37637f6fbe in /nvme1/anaconda3/envs/gyl_mmdetection/lib/python3.7/site-packages/torch/lib/libc10_cuda.so) frame #2: c10::TensorImpl::release_resources() + 0x4d (0x7f37635a3abd in /nvme1/anaconda3/envs/gyl_mmdetection/lib/python3.7/site-packages/torch/lib/libc10.so) frame #3: + 0x523542 (0x7f373883c542 in /nvme1/anaconda3/envs/gyl_mmdetection/lib/python3.7/site-packages/torch/lib/libtorch_python.so) frame #4: + 0x5235e6 (0x7f373883c5e6 in /nvme1/anaconda3/envs/gyl_mmdetection/lib/python3.7/site-packages/torch/lib/libtorch_python.so) frame #27: __libc_start_main + 0xe7 (0x7f3778c65c87 in /lib/x86_64-linux-gnu/libc.so.6)

Aborted (core dumped)

YilinGao-SHU avatar May 09 '22 02:05 YilinGao-SHU

Please check the value of the x1, y1, x2, y2 in tl_bboxes = torch.stack([x1, y1, x2, y2], dim=-1) in pycenternet_head.py, line 1204.

Duankaiwen avatar May 09 '22 02:05 Duankaiwen