Akhila Yerukola

icon indicating an email [email protected]

icon location Stanford icon location Stanford CS Masters '19 | Researcher at Stanford NLP | NIT Trichy, India CSE '16

Results 2 comments of Akhila Yerukola

when running train_discriminator.py ,occuring cudnn RNN backward can only be called in training mode

Replace `x_gen_attr` with `x_gen_attr.detach()` in `train_discriminator.py`. i.e in `model.forward_encoder_embed` and `model.forward_discriminator_embed`

Why is my loss in the verification set going up but my loss in the training set going down

It seems that you're overfitting on your train set. Either add regularization or do early stopping. There are lots of other ways to prevent overfitting as well.