- 20 Jul, 2020 14 commits
-
-
Sam Shleifer authored
Huge MT speedup!
-
Sylvain Gugger authored
* Improve doc of use_cache * Update src/transformers/configuration_xlnet.py Co-authored-by:
Teven <teven.lescao@gmail.com> Co-authored-by:
Teven <teven.lescao@gmail.com>
-
Clement authored
-
Clement authored
-
Clement authored
-
Clement authored
-
Clement authored
-
Sam Shleifer authored
-
Stas Bekman authored
* DataParallel fixes: 1. switched to a more precise check - if self.args.n_gpu > 1: + if isinstance(model, nn.DataParallel): 2. fix tests - require the same fixup under DataParallel as the training module * another fix
-
Pradhy729 authored
* Don't pass sampler for iterable dataset * Added check for test and eval dataloaders. * Formatting * Don't pass sampler for iterable dataset * Added check for test and eval dataloaders. * Formatting * Cleaner if nesting. * Added test for trainer and iterable dataset * Formatting for test * Fixed import when torch is available only. * Added require torch decorator to helper class * Moved dataset class inside unittest * Removed nested if and changed model in test * Checking torch availability for IterableDataset
-
Julien Chaumond authored
cc @lhoestq @thomwolf Also cc'ing model author @nreimers => Model pages now properly link to the dataset pages (and in the future, eval results, etc.)
-
Manuel Romero authored
-
Alan deLevie authored
-
Alan deLevie authored
-
- 18 Jul, 2020 8 commits
-
-
Sam Shleifer authored
Co-authored-by:Pradhy729 <49659913+Pradhy729@users.noreply.github.com>
-
Teven authored
Slightly breaking change, changes functionality for `use_cache` in XLNet: if use_cache is True and mem_len is 0 or None (which is the case in the base model config), the model behaves like GPT-2 and returns mems to be used as past in generation. At training time `use_cache` is overriden and always True.
-
Teven authored
Slightly breaking change, changes functionality for `use_cache` in XLNet: if use_cache is True and mem_len is 0 or None (which is the case in the base model config), the model behaves like GPT-2 and returns mems to be used as past in generation. At training time `use_cache` is overriden and always True.
-
Sebastian authored
-
Sam Shleifer authored
-
Sam Shleifer authored
-
Nathan Raw authored
Co-authored-by:Sam Shleifer <sshleifer@gmail.com>
-
- 17 Jul, 2020 11 commits
-
-
Teven authored
Slightly breaking change, changes functionality for `use_cache` in XLNet: if use_cache is True and mem_len is 0 or None (which is the case in the base model config), the model behaves like GPT-2 and returns mems to be used as past in generation. At training time `use_cache` is overriden and always True.
-
Jannes authored
-
Julien Chaumond authored
-
Gianpaolo Di Pietro authored
* Added model card for neuraly/bert-base-italian-cased-sentiment * Update model_cards/neuraly/bert-base-italian-cased-sentiment/README.md Co-authored-by:
Julien Chaumond <chaumond@gmail.com> Co-authored-by:
Gianpy15 <g.dipietro@neuraly.ai> Co-authored-by:
Julien Chaumond <chaumond@gmail.com>
-
Patrick von Platen authored
Add Rouge2 results
-
Patrick von Platen authored
* fix merge rebase * add intermediate reformer code * save intermediate caching results * save intermediate * save intermediate results * save intermediate * upload next step * fix generate tests * make tests work * add named tuple output * Apply suggestions from code review * fix use_cache for False case * fix tensor to gpu * fix tensor to gpu * refactor * refactor and make style
-
Patrick von Platen authored
* Create README.md * Update README.md * Update README.md * Update README.md
-
Sam Shleifer authored
-
Sam Shleifer authored
-
Bayartsogt Yadamsuren authored
* language tag addition on albert-mongolian * Update model_cards/bayartsogt/albert-mongolian/README.md Co-authored-by:Julien Chaumond <chaumond@gmail.com>
- 16 Jul, 2020 7 commits
-
-
Manuel Romero authored
-
Nick Doiron authored
-
Sam Shleifer authored
-
Manuel Romero authored
Fix missig "-" in meta data
-
Patrick von Platen authored
-
Patrick von Platen authored
-
Patrick von Platen authored
-