GitLab

Fixes flake8 warning W291 (x224).

Submodules shouldn't import from their parent in general.

We need https://github.com/timothycrosley/isort/pull/1000 but there's no
release with this fix yet, so we'll install from GitHub.

This is the result of:

    $ isort --recursive examples templates transformers utils hubconf.py setup.py

lines_after_imports = 2 is a matter of taste; I like it.

This is the result of:

    $ black --line-length 119 examples templates transformers utils hubconf.py setup.py

There's a lot of fairly long lines in the project. As a consequence, I'm
picking the longest widely accepted line length, 119 characters.

This is also Thomas' preference, because it allows for explicit variable
names, to make the code easier to understand.

Likely it was added by accident.

adding positional embeds masking to TFRoBERTa

[WIP] Add MMBT Model to Transformers Repo

closes #1960 Add saving and resuming functionality for remaining examples

[BREAKING CHANGE] Setting all ignored index to the PyTorch standard

run_squad with roberta

fix: wrong architecture count in README

:zip: #2106 tokenizer.tokenize speed improvement (3-8x) by caching added_tokens in a Set

[WIP] Sampling sequence generator for transformers

fix run_squad.py during fine-tuning xlnet on squad2.0

Fix TFXLNet

Add special tokens to documentation for bert examples to resolve issue: #1561

Bug-fix: Roberta Embeddings Not Masked

Support running tests in parallel

This looks like debug code accidentally committed in b18509c2.

Refs #2250.