pytorch-faster-rcnn
pytorch-faster-rcnn copied to clipboard
RuntimeError: merge_sort: failed to synchronize: an illegal memory access was encountered.When train by myself with "./experiments/scripts/train_faster_rcnn.sh 0 pascal_voc vgg16"
After Loading initial model weights from data/imagenet_weights/vgg16.pth Loaded. It will appear this problem:(,(I didn't change anything,just try to train it)
File "/home/hdd2/pytorch-faster-rcnn/tools/../lib/nets/network.py", line 87, in _proposal_layer self._feat_stride, self._anchors, self._num_anchors) File "/home/hdd2/pytorch-faster-rcnn/tools/../lib/layer_utils/proposal_layer.py", line 35, in proposal_layer scores, order = scores.view(-1).sort(descending=True) RuntimeError: merge_sort: failed to synchronize: an illegal memory access was encountered Command exited with non-zero status 1 15.43user 4.00system 0:19.74elapsed 98%CPU (0avgtext+0avgdata 4566528maxresident)k 0inputs+32outputs (0major+1414420minor)pagefaults 0swaps
Solved?
It seems that change pytorch version to 0.4.1 is ok, and got problems on pytorch 0.3.1 and pytorch 1.0. On 1.1, the error is "RuntimeError: merge_sort: failed to synchronize: an illegal memory access was encountered"
After Loading initial model weights from data/imagenet_weights/vgg16.pth Loaded. It will appear this problem:(,(I didn't change anything,just try to train it)
File "/home/hdd2/pytorch-faster-rcnn/tools/../lib/nets/network.py", line 87, in _proposal_layer self._feat_stride, self._anchors, self._num_anchors) File "/home/hdd2/pytorch-faster-rcnn/tools/../lib/layer_utils/proposal_layer.py", line 35, in proposal_layer scores, order = scores.view(-1).sort(descending=True) RuntimeError: merge_sort: failed to synchronize: an illegal memory access was encountered Command exited with non-zero status 1 15.43user 4.00system 0:19.74elapsed 98%CPU (0avgtext+0avgdata 4566528maxresident)k 0inputs+32outputs (0major+1414420minor)pagefaults 0swaps
Have you solved this issue?