how much GPU memory does the next QA need?

Question

how much GPU memory does the next QA need?

Opened this issue 2 years ago · 6 comments

Answer 1 · 2022-05-12T02:45:41.000Z

Thanks for the question. It needs about 24G for training of the model with batch size 64 with 8 clips for each video, whereas 8G is enough for inference. If you want to train with 16 clips, you need to change batchsize to 32.

Answer 2 · 2022-05-12T02:56:24.000Z

Thank you for your reply.

Answer 3 · 2022-05-14T02:15:46.000Z

Well，Could you offer the pre-trained bert model？ Thanks.

Answer 4 · 2022-05-16T07:52:02.000Z

Well, I also need this

Answer 5 · 2022-05-16T08:42:48.000Z

Hi, please find the code and mode for BERT finetuning/feature extraction. You should launch a new issue with a proper title, otherwise I may slow in finding your questions..

Answer 6 · 2022-05-16T11:24:44.000Z

ok, many thanks. Is the model in nextqa for question features the final model? Or I still need to train? I just utilize the model to extract question feature while the results drop significantly by nearly 5 points.