GPU out of memory error

1

“CUDA out of memory error”
Getting this error here test data is an iterator.
How to fix it?
Here is the code where error is occuring?
Can i take batches and test?

get predictions for test data

with torch.no_grad():
preds = model(test_seq.to(device), test_mask.to(device))
preds = preds.detach().cpu().numpy()

2

Your batch size might be too large, so you could try to lower it during the test run.

PS: you can post code snippets by wrapping them into three backticks ``` :wink: