windspirit95

> Hi! @ethman I have reproduced your indicated steps as follows: > > ``` > model_path = nussl.efz_utils.download_trained_model('mask-inference-wsj2mix-model-v1.pth') > > # workaround for known issue in the library > #...

> I can run train on my GPU with batch size = 24, but on validation I've got OOM, nevertheless on validation I don't need gradients, only weights, so validation...

I have tried on timsler 's finetune notebook, and actually it works well on my dataset. (~90% accuracy with 49 classes) Something to be noted in my experience: - The...