About copying a param with shape torch.Size([13380,468]) from checkpoint,the shape in current model is torch.Size([13475,468])])

[maoao686868]

@hobincar
Hello, when we download pretrained Recnet_Global model and pretreated features from MSVD dataset to python run.py. I come across this problem:
Traceback (most recent call last):
File "run.py", line 96, in
run('/root/Workspace/rn/checkpoint/RecNet-global_MSVD(1).ckpt')
File "run.py", line 44, in run
decoder.load_state_dict(checkpoint['decoder'])
File "/root/anaconda3/envs/rn/lib/python3.6/site-packages/torch/nn/modules/module.py", line 777, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for Decoder:
size mismatch for embedding.weight: copying a param with shape torch.Size([13380, 468]) from checkpoint, the shape in current model is torch.Size([13475, 468]).
size mismatch for out.weight: copying a param with shape torch.Size([13380, 512]) from checkpoint, the shape in current model is torch.Size([13475, 512]).
size mismatch for out.bias: copying a param with shape torch.Size([13380]) from checkpoint, the shape in current model is torch.Size([13475]).
Please help me, I feel very anxious about this problem.

[maoao686868]

@hobincar
plus.I use environment as follows : using anaconda3 create python3.6, pytorch 1.1,torchvision0.3.0

[hobincar]

Hi. The error message is saying that the model has 13475 vocabs but the checkpoint has 13380 vocabs. I guess the checkpoint is saved using the old implementation. I don't have much time for handling this for now, but I'll look at it. Maybe you can train the model by yourself instead of using my checkpoint file.

[maoao686868]

@hobincar Sorry , this problem is because of my wrong virtual enviroment. I have sovled this problem, thanks for your reply.

[hobincar]

That's great :) Thank you for the information.