RuntimeError: Error(s) in loading state_dict for RNN_ENCODER:

[hs457681503]

size mismatch for encoder.weight: copying a param of torch.Size([1, 300]) from checkpoint, where the shape is torch.Size([5450, 300]) in current model.

[hannanyi]

Look at your birds dataset's caption. Pickle if it's 6M it will run and 4M it will report an error. Here's how mine worked out