search

showlab / all-in-one

[CVPR2023] All in One: Exploring Unified Video-Language Pre-training
https://arxiv.org/abs/2203.07303
277 stars 16 forks source link

downstream task checkpoint #4

Closed fangz-cs closed 2 years ago

fangz-cs commented 2 years ago

Thank you very much for your excellent code sharing, and can you share your downstream tasks checkpoints?

FingerRec commented 2 years ago

Of course. But we take 10+ datasets and produce a large number of downstream checkpoints for each dataset.

The checkpoint for which dataset you need? Maybe I can upload one to google driver if you have difficult in running downstream code.

fangz-cs commented 2 years ago

ok,i just need the VideoQA checkpoint, both open ended and multichoice.Thank you!

FingerRec commented 2 years ago

We upload trained weight on downstream VQA (tgif-frame dataset). The checkpoint is in google driver:

https://drive.google.com/file/d/1yGSjyhcJT0tGURyFT4wa-iVXFOhLP29T/view?usp=sharing

Pls consider star our code if helpful. :)