You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First I run the data preprocessing section and it worked.
Then when I tried to run the Training section ( make train --[args] ...) it shows this error:
`
========== SUMMARY ==========
-> h5 dataset folder path : /path/FUSION-human-action-recognition/data/processed/old_h5_folder_2/
-> output_folder : /path/FUSION-human-action-recognition/models/
-> evaluation_type : cross_subject
-> model_type : FUSION
-> use pose : True
-> use ir : True
-> pretrained : True
-> use cropped ir : True
-> fusion scheme : CONCAT
-> optimizer : ADAM
-> learning rate : 0.0001
-> weight decay : 0.0
-> gradient threshold : 10.0
-> max epochs : 15
-> batch size : 16
-> accumulation steps : 1
-> sub_sequence_length : 20
-> augment_data : True
-> mirror skeleton : False
-> evaluate_test : True
-> seed : 0
Computing c_min and c_max. This takes a while ...
Done !
0 - 0.0/107
Traceback (most recent call last):
File "src/models/train_model.py", line 254, in
validation_generator)
File "/path/FUSION-human-action-recognition/src/models/train_utils.py", line 261, in train_model
out = model(X)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/path/FUSION-human-action-recognition/src/models/pose_ir_fusion.py", line 106, in forward
out_pose = self.pose_net(X_skeleton)[:, :, 0, 0] # shape (batch_size, 512)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/container.py", line 92, in forward
input = module(input)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 343, in forward
return self.conv2d_forward(input, self.weight)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 340, in conv2d_forward
self.padding, self.dilation, self.groups)
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same
make: *** [train] Erreur 1
`
Could you help detect and solve the problem
Thanks in advance for helping !!!!
The text was updated successfully, but these errors were encountered:
Hi !
First I run the data preprocessing section and it worked.
Then when I tried to run the Training section ( make train --[args] ...) it shows this error:
`
========== SUMMARY ==========
-> h5 dataset folder path : /path/FUSION-human-action-recognition/data/processed/old_h5_folder_2/
-> output_folder : /path/FUSION-human-action-recognition/models/
-> evaluation_type : cross_subject
-> model_type : FUSION
-> use pose : True
-> use ir : True
-> pretrained : True
-> use cropped ir : True
-> fusion scheme : CONCAT
-> optimizer : ADAM
-> learning rate : 0.0001
-> weight decay : 0.0
-> gradient threshold : 10.0
-> max epochs : 15
-> batch size : 16
-> accumulation steps : 1
-> sub_sequence_length : 20
-> augment_data : True
-> mirror skeleton : False
-> evaluate_test : True
-> seed : 0
Computing c_min and c_max. This takes a while ...
Done !
0 - 0.0/107
Traceback (most recent call last):
File "src/models/train_model.py", line 254, in
validation_generator)
File "/path/FUSION-human-action-recognition/src/models/train_utils.py", line 261, in train_model
out = model(X)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/path/FUSION-human-action-recognition/src/models/pose_ir_fusion.py", line 106, in forward
out_pose = self.pose_net(X_skeleton)[:, :, 0, 0] # shape (batch_size, 512)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/container.py", line 92, in forward
input = module(input)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/module.py", line 547, in call
result = self.forward(*input, **kwargs)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 343, in forward
return self.conv2d_forward(input, self.weight)
File "/user1/path/.conda/envs/fusion/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 340, in conv2d_forward
self.padding, self.dilation, self.groups)
RuntimeError: Input type (torch.FloatTensor) and weight type (torch.cuda.FloatTensor) should be the same
make: *** [train] Erreur 1
`
Could you help detect and solve the problem
Thanks in advance for helping !!!!
The text was updated successfully, but these errors were encountered: