ebsmothers
ebsmothers
Hi and thanks for using MMF. Can you reach out to the owner of OKVQA? They are more likely to be able to help you with this question.
> CUDA_VISIBLE_DEVICES=0,3,6,7 👀
> Hello guys, what else do we need here? Can I add these changes to the other recipes? Sorry for the delay in responding. For the sake of not writing...
> I am not sure the GPU metrics is something you want to log very often as it is expensive to compute. We could consider adding a flag like `log_memory_stats`...
Thanks for an amazing analysis and for writing up the entire process so clearly! A few comments and questions (some may be pretty basic, I need to read through this...
Thanks for the PR @albanD! Tbh we have already had a fraught relationship with meta device initialization 😅 (see e.g. #317, #418, #514). Our latest status is that we deliberately...
Hi @bhack thanks for the question! We haven't added any multimodal models yet as we are working to get good coverage of text-only methods first, but it's definitely something we...
Thanks @bhack and @matbee-eth for the suggestions! At this exact moment we do not have the bandwidth to take these on but we will keep them both in mind for...
Thanks @yechenzhi for opening the PR! Sorry for the delay here, this one slipped through the cracks over the excitement of the past couple days. I will take a proper...
> The loss appeared normal; however, I encountered an issue where utilizing two GPUs did not expedite the training process. I also attempted the 'tune run --nnodes 1 --nproc_per_node 2...