sgugger

Follow

Sylvain Gugger sgugger

Follow

Research Engineer at HuggingFace

2.1k followers · 4 following

Achievements

BetaSend feedback

Achievements

BetaSend feedback

Organizations

Block or Report

Block or report sgugger

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Popular repositories

Deep-Learning Public

A few notebooks about deep learning in pytorch

Jupyter Notebook 530 111
Adam-experiments Public

Experiments with Adam/AdamW/amsgrad

Python 178 36
hf_examples Public

NLP Examples using the 🤗 libraries

Jupyter Notebook 34 4
SwiftData Public

Swift 15 1
fastai Public

Forked from fastai/fastai

The fast.ai deep learning library, lessons, and tutorials

Jupyter Notebook 7 2
transformer_train Public

Python 7

3,365 contributions in the last year

Learn how we count contributions

Less More

NEW! View your contributions in 3D, VR and IRL!

Activity overview

Contributed to huggingface/transformers, huggingface/accelerate, huggingface/doc-builder and 20 other repositories

Contribution activity

August 2022

Created 42 commits in 4 repositories

Created a pull request in huggingface/huggingface_hub that received 12 comments

[PoC] Cache non-existence of files or completeness of repo

In Transformers, we very often try to download files that do not exist: this is because each tokenizer has a list of optional files so we try to do…

+105 −2 • 12 comments

Opened 18 other pull requests in 4 repositories

huggingface/transformers 13 merged

Fix docstrings with last version of hf-doc-builder styler Aug 11
Properly move cache when it is not in default path Aug 10
Preserve hub-related kwargs in AutoModel.from_pretrained Aug 9
Use commit hash to look in cache instead of calling head Aug 8
New cache fixes: add safeguard before looking in folders Aug 8
Clean up hub Aug 5
Move cache folder to huggingface/hub for consistency with hf_hub Aug 5
Fix pipeline tests Aug 5
Remove py.typed Aug 5
Fix load of model checkpoints in the Trainer Aug 4
Use new huggingface_hub tools for download models Aug 2
Fix ROUGE add example check and update README Aug 1
Add evaluate to test dependencies Aug 1

huggingface/doc-builder 2 merged

Fix bug in styler when line after Parameter block is empty Aug 9
Fix tests Aug 2

huggingface/huggingface_hub 2 merged

huggingface/accelerate 1 merged

Fix DispatchDataloader Aug 1

Reviewed 130 pull requests in 6 repositories

huggingface/transformers 25 pull requests

Generate: validate model_kwargs (and catch typos in generate arguments) Aug 11
fix owlvit tests, update docstring examples Aug 11
Fix failure on DeBERTa(base/v2/sew_d) fp16 training with ONNX Runtime Aug 11
Add checks for some workflow jobs Aug 11
Add Donut Aug 11
Bump nbconvert from 6.0.1 to 6.3.0 in /examples/research_projects/visual_bert Aug 11
Bump nbconvert from 6.0.1 to 6.3.0 in /examples/research_projects/lxmert Aug 11
Load sharded pt to flax Aug 11
[FX] _generate_dummy_input supports audio-classification models for labels Aug 11
Deberta V2: Fix critical trace warnings to allow ONNX export Aug 11
Return the permuted hidden states if return_dict=True Aug 11
german docs translation Aug 11
Change BartLearnedPositionalEmbedding's forward method signature to support Opacus training Aug 11
Update BLOOM parameter counts Aug 11
Supporting seq2seq models for bitsandbytes integration Aug 11
update doc for perf_train_cpu_many, add intel mpi introduction Aug 11
Fix LayoutLMv3 documentation Aug 11
The backticks in the example of transformers.BigBirdPegasusConfig documentation were not in the right spot… Aug 11
Adds CLIP to models exportable with ONNX Aug 10
pipeline support for device="mps" (or any other string) Aug 10
raise atol for MT5OnnxConfig Aug 10
Use commit hash to look in cache instead of calling head Aug 10
TF Examples Rewrite Aug 10
Run torch_fx tests in a spawn process to avoid memory issue Aug 10
Update philosophy to include other preprocessing classes Aug 10
Some pull request reviews not shown.

huggingface/accelerate 25 pull requests

Fix multi-node issues and simplify param logic Aug 11
launcher related changes + minor fixes Aug 11
M1 mps fixes Aug 11
Fully remove subprocess from the multi-gpu launcher Aug 10
Integrate Rich into Accelerate Aug 10
Fix grad accum example in docs Aug 9
Added GANs example to examples Aug 9
Make running tests more efficient Aug 8
Fix DeepSpeed CI Aug 8
Fix typo in docs/index.mdx Aug 8
Make send_to_device() run asynchronously Aug 5
Correct dtype casting Aug 4
fixing deepspeed slow tests issue Aug 4
Fix some typos + wordings Aug 3
reorg of test scripts and minor changes to tests Aug 3
M1 GPU mps device integration Aug 3
Pass in config_kwargs to init Aug 2
Improve warning Aug 2
Reenable Gather for Metrics Aug 2
Pin deepspeed Aug 2
Shorthand way to grab a tracker Aug 2
Improve docstring for on_local_process Aug 1
TESTS! - Oops Aug 1
Fix DispatchDataloader Aug 1
Complete revamp of the docs Aug 1

huggingface/huggingface_hub 4 pull requests

huggingface/notebooks 3 pull requests

amyeroberts/transformers 3 pull requests

GLPN - Rename featue extractor to image processor glpn Aug 8
Image processor glpn Aug 8
Image processor mixin Aug 8

huggingface/doc-builder 1 pull request

Adapt index for namespace packages in autodoc.get_shortest_path Aug 2

Created an issue in huggingface/huggingface_hub that received 8 comments

Use a finer exception when local_files_only=True and a file is missing in cache

In Transformers we sometime try to access files that are not in a repo, catch the error (EntryNotFoundError for distant repos) and try a different …

8 comments

Opened 2 other issues in 1 repository

huggingface/huggingface_hub 2 open

Add ability to turn on and off progress bars Aug 9
Add a utility to list cached things Aug 9