Gueter Josmy Faure comments

Results 6 comments of


                                            Gueter Josmy Faure

Distributed training problem

I have not encountered this issue. If you're training JHMDB, one GPU is ok, it should not take long to train.

Could MultiSports Dataset be tranformed to like ucf101-24 dataset or JHMDB dataset?

It could. To get started with Multisports, I would suggest the other repository HIT_ava. The code does not follow the methodology of the published paper but it's simpler and more...

Could MultiSports Dataset be tranformed to like ucf101-24 dataset or JHMDB dataset?

I did not skip any frame for AVA (ran the code on video data) For Multisports, I use all the frames provided by the authors of the dataset.

Misror error in window on pip install-e

I did not try the program in windows, so I can't answer that. I don't remember needing ninja for linux

torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

Decreasing the batch size worked for me

Error with batch inference

What is your `max_new_tokens`? I find that for batch inference that value should not be too large. For me 64 works in most cases (but the answer might be truncated)