VILA icon indicating copy to clipboard operation
VILA copied to clipboard
verified publisher icon NVlabs

Is there any DataCollatorForSupervisedDatasetSeqParallel for video dataset?

Open ymbao97 opened this issue 9 months ago • 1 comments

The current DataCollatorForSupervisedDatasetSeqParallel in llava/data/dataset.py is built for image dataset. There will be many errors when directly using it for video dataset. Will you release the similar solution for video training, so as to enbale re-implementation of stage 4 and 5 in LongVILA?

ymbao97 avatar Feb 26 '25 09:02 ymbao97

@yukang2017 could you help provide a pointer to the SeqParallel dataset?

Lyken17 avatar Feb 28 '25 00:02 Lyken17