GitLab

PyTorch < 1.3 requires multiplication operands to be of the same type.
This was violated when using default attention mask (i.e.,
attention_mask=None in arguments) given BERT in the decoder mode.

In particular, this was breaking Model2Model and made tutorial
from the quickstart failing.

* [model_cards] New German Europeana BERT models from dbmdz

* [model_cards] Update German Europeana BERT models from dbmdz

Fix circleci cuInit error on Tensorflow >= 2.1.0.

This will enable filtering on language (amongst other tags) on the website

cc @loretoparisi, @stefan-it, @HenrykBorzymowski, @marma

* Create README.md

* Update README.md

* Update README.md

* Update README.md

* [model_cards] Use code fences for consistency
Co-authored-by: Julien Chaumond <chaumond@gmail.com>

* Add readme for deepset/roberta-base-squad2

* update model readme

cc @jetrunner
Co-Authored-By: Kevin Canwen Xu <canwenxu@outlook.com>

cc @tholor

see #2799

* add readme for bert-base-german-cased

* update readme

Add option to `cached_path` to automatically extract archives

Tensorflow 2.1.0 introduce a new dependency model where pip install tensorflow would install tf with GPU support.
Before it would just install with CPU support, thus CircleCI is looking for NVidia driver version at initialization of the
tensorflow related tests but fails as their is no NVidia Driver running.
Signed-off-by: Morgan Funtowicz <morgan@huggingface.co>

Changed vocabulary save function. Variable name was inconsistent, causing an error to be thrown when passing a file name instead of a directory.