Simpletransformers

Added

- Added BERTweet for multilabel classification [manueltonneau](https://github.com/manueltonneau)

Fixed

- Fixed bug where `T5Model` would save evaluation checkpoints even when `save_eval_checkpoints` is False.
- Fixed bug where `args.silent` was not used in `NERModel`. [mossadhelali](https://github.com/mossadhelali)

Changed

- Changed the default value of `dataloader_num_workers` (for Pytorch Dataloaders) to 0. This is to avoid memory leak issues with Pytorch multiprocessing with text data.

Added

- Added support for MT5
- Added support for Adafactor optimizer
- Added support for various schedulers:
- get_constant_schedule
- get_constant_schedule_with_warmup
- get_linear_schedule_with_warmup
- get_cosine_schedule_with_warmup
- get_cosine_with_hard_restarts_schedule_with_warmup
- get_polynomial_decay_schedule_with_warmup

Changed

- `T5Model` now has a required `model_type` parameter (`"t5"` or `"mt5"`)

Fixed

- Fixed issue with class weights not working in `ClassificationModel` when using mult-GPU training