Size mismatch issue while testing retrained ESPNet on Cityscapes

[prajakta-13]

Hi Sachin,
Thank you for your work!
I am trying to use the repo and retrain ESPNet on Cityscapes. I have generated a decoder model and while trying to generate inference on the same, I am facing a runtime error depicting size mismatch.

size mismatch for conv.conv.weight: copying a param of torch.Size([20, 36, 3, 3]) from checkpoint, where the shape is torch.Size([20, 39, 3, 3]) in current model.

Can you please let me know what might be going wrong here?

Thanks in advance!
_Prajakta Kulkarni

[sacmehta]

See #14