Shape of check point and shape from in current modal are not same

vision
1

I am trying to load a pre-trained modal but for some reason, the dimensions are not setting current.

  • size mismatch for module.Prediction.weight: copying a param with shape torch.Size([50, 256]) from checkpoint, the shape in current model is torch.Size([37, 256]).
  • size mismatch for module.Prediction.bias: copying a param with shape torch.Size([50]) from checkpoint, the shape in current model is torch.Size([37]).

Can someone tell me what’s happening?

2

Can you send the code for loading the models.

3

I got it the input size passing to the trained model was changed.

1 Like
4

Ok great glad you solved it. Just mark your answer as solved so people don’t think you need more help.

1 Like