size mismatch for my video

[fengjingchehu]

When I try to visualize the video, I encounter a size mismatch issue, which is also encountered in Alphaction. So how can I solve this problem?

Traceback (most recent call last): File "demo.py", line 99, in main() File "demo.py", line 46, in main checkpointer.load(cfg.MODEL.WEIGHT) File "/mnt/745425dc-5d5b-4490-9d38-57169dcb2ab3/wxl/HIT-master/hit/utils/checkpoint.py", line 61, in load self._load_model(checkpoint, no_head) File "/mnt/745425dc-5d5b-4490-9d38-57169dcb2ab3/wxl/HIT-master/hit/utils/checkpoint.py", line 110, in _load_model load_state_dict(self.model, checkpoint.pop("model"), no_head) File "/mnt/745425dc-5d5b-4490-9d38-57169dcb2ab3/wxl/HIT-master/hit/utils/model_serialization.py", line 83, in load_state_dict model.load_state_dict(model_state_dict) File "/home/user/anaconda3/envs/hit/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1406, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for ActionDetector: size mismatch for roi_heads.action.feature_extractor.hit_structure.person_dim_reduce.weight: copying a param with shape torch.Size([1024, 2304, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([1024, 256, 1, 1, 1]). size mismatch for roi_heads.action.feature_extractor.hit_structure.person_key_dim_reduce.weight: copying a param with shape torch.Size([1024, 2304, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([1024, 256, 1, 1, 1]). size mismatch for roi_heads.action.feature_extractor.hit_structure.object_dim_reduce.weight: copying a param with shape torch.Size([1024, 2304, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([1024, 256, 1, 1, 1]). size mismatch for roi_heads.action.feature_extractor.hit_structure.hand_dim_reduce.weight: copying a param with shape torch.Size([1024, 2304, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([1024, 256, 1, 1, 1]). size mismatch for roi_heads.action.feature_extractor.fc1.weight: copying a param with shape torch.Size([1024, 2304]) from checkpoint, the shape in current model is torch.Size([1024, 256]).