microsoft / GenerativeImage2Text

GIT: A Generative Image-to-text Transformer for Vision and Language
MIT License
546 stars 68 forks source link

Where can we download pretrained weights for Git MSVD-QA? #57

Closed ee2110 closed 3 weeks ago

ee2110 commented 11 months ago

Hi, thank you so much for the great work and releasing the code. I would like to study on videoQA ability of this model, specifically or MSVD-QA or TGIF-Frame, is that possible for us to download the fine-tuned or pre-trained weights so that we could use them for zero-shot on datasets? Thank you.

hanranCode commented 9 months ago

you can find model from huggingface. eg: “git-base-vqav2” model -> https://huggingface.co/microsoft/git-base-vqav2

amsword commented 3 weeks ago

please re-open it for any other question.