Add TF weights

#1
by joaogante HF staff - opened

Model converted by the transformers' pt_to_tf CLI.

All converted model outputs and hidden layers were validated against its Pytorch counterpart. Maximum crossload output difference=7.504e-03; Maximum converted output difference=7.504e-03.

Related PR: https://github.com/huggingface/transformers/pull/17554

Now converting from shards to shards! cc @sayakpaul @nielsr

My god. Can't believe we have the largest TF checkpoint in vision now!

(merging as agreed on Slack)

joaogante changed pull request status to merged

Sign up or log in to comment