- 06 Oct, 2021 12 commits
-
-
Lysandre authored
-
Anton Lozhkov authored
-
Sylvain Gugger authored
-
Yanming Wang authored
* Fix logging_nan_inf_filter in torch_xla mode * Update src/transformers/trainer.py Co-authored-by:
Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Fix format Co-authored-by:
-
yssjtu authored
* enabling using past_key_values together with labels when training in T5ForConditionalGeneration * test * Enable past_key_values in T5ForconditionalGeneration while training. * delete comments
-
Akul Agrawal authored
-
Nicolas Patry authored
Fixes #13846
-
David del Río Medina authored
-
Md Saiful Islam Sayef authored
* Add example use of few-shot named entity recognition model in research_projects folder. * Apply suggestions from code review Co-authored-by:
NielsRogge <48327001+NielsRogge@users.noreply.github.com> * Update fsner example README.md. - Change wrong import FSNERTokenizerWrapper to FSNERTokenizerUtils in the example code - Add a link to the model identifier * Update examples/research_projects/fsner/src/fsner/model.py Fix spelling mistake in the default parameter of pretrained model name. Co-authored-by:
Stefan Schweter <stefan@schweter.it> * Add example use of few-shot named entity recognition model in research_projects folder. * Apply suggestions from code review Co-authored-by:
-
Nicolas Patry authored
Co-authored-by:
Pierre Snell <pierre.snell@botpress.com> Co-authored-by:
-
Sylvain Gugger authored
-
Hyunwoong Ko authored
* Update parallelism.md * Update docs/source/parallelism.md Co-authored-by:
Stas Bekman <stas00@users.noreply.github.com> * Update docs/source/parallelism.md Co-authored-by:
-
- 05 Oct, 2021 10 commits
-
-
Siarhei Melnik authored
-
Boris Dayma authored
-
Nicolas Patry authored
* Tmp. * Fixing BC for question answering with long context. * Capping model_max_length to avoid tf overflow. * Bad workaround bugged roberta. * Fixing name.
-
Zhaofeng Wu authored
* Allow dataset to be an optional argument for (Distributed)LengthGroupedSampler * Fix
-
Michael Benayoun authored
* Symbolic trace dynamic axes support for BERT like models (albert, bert, distilbert, mobilebert, electra, megatron-bert) * Sanity checks before tracing that make sure the model to trace is supported * Adapted to PyTorch 1.9 Co-authored-by:Michael Benayoun <michael@huggingface.co>
-
Alex Hedges authored
* Improve error message when loading models from Hub * Adjust error message wording
-
Nicolas Patry authored
* Fixing empty prompts for text-generation when BOS exists. * Fixing odd case with Pegasus. * Fixing Bert is Assertion Error.
-
Yih-Dar authored
Co-authored-by:ydshieh <ydshieh@users.noreply.github.com>
-
Nicolas Patry authored
-
Sam Hardwick authored
* Update Tatoeba conversion
-
- 04 Oct, 2021 6 commits
-
-
Bram Vanroy authored
* update no_* argument Changes the order so that the no_* argument is created after the original argument AND sets the default for this no_* argument to False * import copy * update test * make style * Use kwargs to set default=False * make style
-
Nathan Raw authored
*
✨ update image classification example *📌 update reqs -
Evgeniy Zheltonozhskiy authored
* Fix broken link to distill models * Missing symbol * Fix spaces
-
Sidd Karamcheti authored
* Add layer-wise scaling * Add reorder & upcasting argument * Add OpenAI GPT-2 weight initialization scheme * start `layer_idx` count at zero for consistency * disentangle attn and reordered and upscaled attn function * rename `scale_attn_by_layer` to `scale_attn_by_layer_id` * make autocast from amp compatible with pytorch<1.6 * fix docstring * style fixes * Add fixes from PR feedback, style tweaks * Fix doc whitespace * Reformat * First pass scale_attn_by_layer_idx and reorder_and_upcast_attn tests * Rename scale_attn_by_layer_idx, add tip * Remove extra newline * add test for weight initialization * update code format * add assert check weights are fp32 * remove assert * Fix incorrect merge * Fix shape mismatch in baddbmm * Add generation test for Mistral flags Co-authored-by:
leandro <leandro.vonwerra@spoud.io> Co-authored-by:
Keshav Santhanam <keshav2@stanford.edu> Co-authored-by:
J38 <jebolton@stanford.edu>
-
Yaser Abdelaziz authored
-
Gunjan Chhablani authored
-
- 01 Oct, 2021 5 commits
-
-
Stas Bekman authored
-
Silviu Oprea authored
In BartForConditionalGeneration.forward, if labels are provided, decoder_input_ids are set to the labels shifted to the right. This is problematic: if decoder_inputs_embeds is also set, the call to self.model, which eventually gets to BartDecoder.forward, will raise an error. The fix is quite simple, similar to what is there already in BartModel.forward. Mainly, we should not compute decoder_input_ids if decoder_inputs_embeds is provided. Co-authored-by:Silviu Vlad Oprea <silviuvo@amazon.co.uk>
-
Anton Lozhkov authored
* Restore broken merge * Additional args, DDP, remove CommonLanguage * Update examples for V100, add training results * Style * Apply suggestions from code review Co-authored-by:
-
Arfon Smith authored
-
Yuta Hayashibe authored
* Removed wrong warning * Raise a warning when `max_length` is given with wrong `truncation` * Update the error message * Update the warning message Co-authored-by:
-
- 30 Sep, 2021 7 commits
-
-
Suraj Patil authored
-
Patrick von Platen authored
* update * add to docs and init * make fix-copies
-
Patrick von Platen authored
-
Gunjan Chhablani authored
* Init multibert checkpoint conversion script * Rename conversion script * Fix MultiBerts Conversion Script * Apply suggestions from code review Co-authored-by:
Patrick von Platen <patrick.v.platen@gmail.com> Co-authored-by:
-
Stas Bekman authored
-
Sylvain Gugger authored
-
Sylvain Gugger authored
-
