GitLab

* Document TF modeling utils

* Document all model utils

improve unittests for finetuning, especially w.r.t testing frozen parameters
fix freeze_embeds for T5
add streamlit setup.cfg

* finish benchmark

* fix isort

* fix setup cfg

* retab

* fix time measuring of tf graph mode

* fix tf cuda

* clean code

* better error message

* Fix: unpin flake8 and fix cs errors

* Ok we still need to quote those

Co-Authored-By: Stefan Schweter <stefan@schweter.it>

* add known 3rd party to setup.cfg

* comment

* Update CONTRIBUTING.md
Co-authored-by: Julien Chaumond <chaumond@gmail.com>

* initial pytorch lightning commit

* tested multigpu

* Fix learning rate schedule

* black formatting

* fix flake8

* isort

* isort

* .
Co-authored-by: Check your git settings! <chris@chris-laptop>

Take advantage of this to simplify the Circle CI configuration.

Don't bother with tensorboardX: it's a fallback for PyTorch < 1.1.0.

These libraries aren't always installed in the virtual environment where
isort is running. Declaring them properly avoids mixing these
third-party imports with local imports.

I don't want to consider it a dependency of transformers, but it's
usually there in local development and usually not there in CI.

I'll fix it later.

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