(ArcFace-Multiplex-Recognition) bc@bcl:~/Projects/Activity-Preserving-Video-Face-Anonymization/demo$ python3 demo_cam.py ==> Face detector loaded initialization method [normal] ==> loading the model from pretrained/model_1.pth ==> netG loaded Traceback (most recent call last): File "demo_cam.py", line 136, in <module> demo(netF, netG) File "demo_cam.py", line 67, in demo box_info = detect(img, netF) File "demo_cam.py", line 27, in detect ssh_rois = netF(img_data, img_info) File "/home/bc/.venv/ArcFace-Multiplex-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/bc/Projects/Activity-Preserving-Video-Face-Anonymization/demo/SSH/model/SSH.py", line 159, in forward m3_ssh_cls_score_reshape = self.reshape(m3_ssh_cls_score, 2) File "/home/bc/Projects/Activity-Preserving-Video-Face-Anonymization/demo/SSH/model/SSH.py", line 51, in reshape input_shape[3] RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.
MyraBaba
commented
3 years ago
Hi @blacknwhite5
I have below error
(ArcFace-Multiplex-Recognition) bc@bcl:~/Projects/Activity-Preserving-Video-Face-Anonymization/demo$ python3 demo_cam.py ==> Face detector loaded initialization method [normal] ==> loading the model from pretrained/model_1.pth ==> netG loaded Traceback (most recent call last): File "demo_cam.py", line 136, in <module> demo(netF, netG) File "demo_cam.py", line 67, in demo box_info = detect(img, netF) File "demo_cam.py", line 27, in detect ssh_rois = netF(img_data, img_info) File "/home/bc/.venv/ArcFace-Multiplex-Recognition/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/bc/Projects/Activity-Preserving-Video-Face-Anonymization/demo/SSH/model/SSH.py", line 159, in forward m3_ssh_cls_score_reshape = self.reshape(m3_ssh_cls_score, 2) File "/home/bc/Projects/Activity-Preserving-Video-Face-Anonymization/demo/SSH/model/SSH.py", line 51, in reshape input_shape[3] RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.