Closed wang9danzuishuai closed 1 year ago
I have the same request, only image type input is supported in the current training pipeline code, but I want to test training this model with data containing video. Thank you very much! I will be very grateful if you can provide code!
same request
No, we do not use any video-instruction data. The video training code can be modified from image version.
Hi! Thank you for your great work. Recently we are preparing for the video Q&A dataset. We want to use mPLUG-Owl to do some experiments. So will there be a release of video training code and dataset instruction like the image-version? Thank you very much! :D