Gabriel Oliveira

I can't train model using 2 GPUs

4

I am trying to train the captioning base model on two Quadro RTX 8000 GPUs, each one with 48GiB RAM. But when I run the command to train the model...

it makes the eval_utils.py load always the MS COCO dataset, in the line `dataset = eval_kwargs.get('dataset', 'coco')`

empirical_sinkhorn_divergence doesn't have a grad_fn

4

## Describe the bug I am trying to use empirical_sinkhorn_divergence as a loss function in pytorch, but the returned tensor does not have a grad_fn, so the gradient can't be...

Overfit

1

How do I know if my model is overfitted?

Hello, If you have any docker image, could you please make it available? thank you!

Problems with reproduction of results

5

Hi @hadifar ! First of all, thanks for sharing your code, it's a good work! I can't reproduce the results for Biomedical dataset, published in the paper "A Self-Training Approach...