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=4.292e-05; Maximum converted output difference=4.292e-05.

Hi there 👋

I'm a TF maintainer at Hugging Face, and this is your most downloaded model whose weights can be automatically converted into TensorFlow, using our tools. We believe that having TF weights would be of interest to the community, and will further boost the visibility of the model.

I also don't want to be a source of spam! Let me know if you are interested in merging these TF weights, and if you would like me to open PRs with TF weights for other models that you own. Alternatively, if you'd like to have the TF weights but no hub notifications, I can also push the weights using admin privileges 🤗

Hi, @joaogante!

Thank you for this contribution, we really appreciate it. TF weights is an old request from our models' users, so we believe that it will certainly be useful to the community.

Regarding other models, it would be interesting to have TF weights for neuralmind/bert-large-portuguese-cased as well, if possible. :)

Thank you again!

neuralmind changed pull request status to merged

<3

Will open a PR for neuralmind/bert-large-portuguese-cased, if the conversion is successful!

Sadly there are some conversion issues (the output logits have a difference of 1e-4, which is non-negligible), so I won't open a PR for now :(