huggingface/transformers: v4.8.0 Integration with the Hub and Flax/JAX support
Creators
- Thomas Wolf1
- Lysandre Debut2
- Sylvain Gugger
- Patrick von Platen
- Julien Chaumond2
- Stas Bekman3
- Sam Shleifer4
- Victor SANH1
- Manuel Romero
- Funtowicz Morgan5
- Julien Plu2
- Suraj Patil6
- Aymeric Augustin7
- Rémi Louf8
- Stefan Schweter
- Denis
- erenup
- Matt1
- Nicolas Patry1
- Joe Davison2
- Anthony MOI1
- Philipp Schmid1
- Teven5
- Piero Molino
- Grégory Châtel9
- Bram Vanroy10
- Clement1
- Kevin Canwen Xu
- Daniel Stancl11
- Philip May12
- 1. @huggingface
- 2. Hugging Face
- 3. Stasosphere Online Inc.
- 4. @facebookresearch
- 5. HuggingFace
- 6. Huggingface
- 7. @qonto
- 8. Freelance
- 9. DisAItek & Intel AI Innovators
- 10. @UGent
- 11. Seznam.cz
- 12. @t-systems-on-site-services-gmbh
Description
v4.8.0 Integration with the Hub and Flax/JAX support Integration with the Hub
Uploading Transformers object to the Hub has never been easier, with all models, tokenizers and configurations having a revamp push_to_hub()
method as well as a push_to_hub
argument in their save_pretrained()
method. The workflow of this method is changed a bit to be more like git, with a local clone of the repo in a folder of the working directory, to make it easier to apply patches (use use_temp_dir=True
to clone in temporary folders for the same behavior as the experimental API).
The Trainer
also has a push_to_hub
API that you need to activate by passing push_to_hub=True
in its TrainingArguments
. You can control the repo name and organization with push_to_hub_model_id
and push_to_hub_organization
. This will upload the model, its configuration and the associated tokenizer, the TensorBoard runs (if tensorboard
is installed) as well as a draft of model card with the training metrics results on the evaluation set.
- Clean push to hub API #12187 (@sgugger)
Flax/JAX is becoming a fully supported backend of the Transformers library with more models having an implementation in it. BART, CLIP and T5 join the already existing models, find the whole list here.
- [Flax] FlaxAutoModelForSeq2SeqLM #12228 (@patil-suraj)
- [FlaxBart] few small fixes #12247 (@patil-suraj)
- [FlaxClip] fix test from/save pretrained test #12284 (@patil-suraj)
- [Flax] [WIP] allow loading head model with base model weights #12255 (@patil-suraj)
- [Flax] Fix flax test save pretrained #12256 (@patrickvonplaten)
- [Flax] Add jax flax to env command #12251 (@patrickvonplaten)
- add FlaxAutoModelForImageClassification in main init #12298 (@patil-suraj)
- [Flax T5] Fix weight initialization and fix docs #12327 (@patrickvonplaten)
- AutoTokenizer: infer the class from the tokenizer config if possible #12208 (@sgugger)
- update desc for map in all examples #12226 (@bhavitvyamalik)
- Depreciate pythonic Mish and support PyTorch 1.9 version of Mish #12240 (@digantamisra98)
- [t5 doc] make the example work out of the box #12239 (@stas00)
- Better CI feedback #12279 (@LysandreJik)
- Fix for making student ProphetNet for Seq2Seq Distillation #12130 (@vishal-burman)
- [DeepSpeed] don't ignore --adafactor #12257 (@stas00)
- Tensorflow QA example #12252 (@Rocketknight1)
- [tests] reset report_to to none, avoid deprecation warning #12293 (@stas00)
- [trainer + examples] set log level from CLI #12276 (@stas00)
- [tests] multiple improvements #12294 (@stas00)
- Trainer: adjust wandb installation example #12291 (@stefan-it)
- Fix and improve documentation for LEDForConditionalGeneration #12303 (@ionicsolutions)
- [Flax] Main doc for event orga #12305 (@patrickvonplaten)
- [trainer] 2 bug fixes and a rename #12309 (@stas00)
- FlaxBartPretrainedModel -> FlaxBartPreTrainedModel #12313 (@sgugger)
- [docs] performance #12258 (@stas00)
- Add CodeCarbon Integration #12304 (@JetRunner)
- Optimizing away the
fill-mask
pipeline. #12113 (@Narsil) - Add output in a dictionary for TF
generate
method #12139 (@stancld) - Flax summarization script #12230 (@patil-suraj)
- Rewrite ProphetNet to adapt converting ONNX friendly #11981 (@jiafatom)
- Flax T5 #12150 (@vasudevgupta7)
- Add mention of the huggingface_hub methods for offline mode #12320 (@LysandreJik)
- [Flax/JAX] Add how to propose projects markdown #12311 (@patrickvonplaten)
- [TFWav2Vec2] Fix docs #12283 (@chenht2010)
- Add all XxxPreTrainedModel to the main init #12314 (@sgugger)
- Conda build #12323 (@LysandreJik)
- Changed modeling_fx_utils.py to utils/fx.py for clarity #12326 (@michaelbenayoun)
Files
huggingface/transformers-v4.8.0.zip
Files
(10.5 MB)
Name | Size | Download all |
---|---|---|
md5:c251ce2ff0226f68bb30f3ba04627aac
|
10.5 MB | Preview Download |
Additional details
Related works
- Is supplement to
- https://github.com/huggingface/transformers/tree/v4.8.0 (URL)