minieval/pytorch-gken

Official Pytorch implementation of the ACM MM'21 paper: Keyframe Extraction from Motion Capture Sequences with Graph based Deep Reinforcement Learning

pytorch-gken's Issues

Evaluation cause KeyError: 'gconv1.lin.weight'

(py3.7) nd@nd46:~/pytorch-gken$ python eval.py Keyframe.bvh 10

Error message:
Processing BVH file...done! Loading GKEN model...Traceback (most recent call last): File "eval.py", line 41, in <module> agent.load_models() File "/home/nd/pytorch-gken/agent.py", line 167, in load_models self.policy_net.load_state_dict(checkpoint['policy_model_state_dict']) File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1392, in load_state_dict load(self) File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1390, in load load(child, prefix + name + '.') File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1390, in load load(child, prefix + name + '.') File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1387, in load state_dict, prefix, local_metadata, True, missing_keys, unexpected_keys, error_msgs) File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1310, in _load_from_state_dict hook(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) File "/home/nd/anaconda3/envs/py3.7/lib/python3.7/site-packages/torch_geometric/nn/dense/linear.py", line 131, in _lazy_load_hook weight = state_dict[prefix + 'weight'] KeyError: 'gconv1.lin.weight'

any help would be appreciated