huggingface/transformers

Pull Requests Commits

prepare_fa2_from_position_ids function bugfix (#33269)

meliksahturker committed 1 year ago

Verified c50b5675

allow unused input parameters passthrough when chunking in asr pipelines (#33889)

VictorAtIfInsurance committed 1 year ago

Verified a0f4f317

Sum gathered input tokens (#34554)

techkang committed 1 year ago

Verified 4dc1a693

🔴 Mllama: fix base prefix (#34874)

zucchini-nlp committed 1 year ago

Verified 1e492afd

[`Deberta/Deberta-v2`] Refactor code base to support compile, export, and fix LLM (#22105)

ArthurZucker committed 1 year ago

Verified 857d46ca

BLIP: fix generation after hub update (#34876)

zucchini-nlp committed 1 year ago

Verified 098962da

Cache: init empty cache when `use_cache` (#34274)

zucchini-nlp committed 1 year ago

Verified c1a85204

Add safe_globals to resume training on PyTorch 2.6 (#34632)

dvrogozh committed 1 year ago

Verified 1339a14d

Fix: Enable prefill phase key value caching of nemotron/minitron models (#34742)

jeongin601 committed 1 year ago

Verified 318fe25f

Fix support for image processors modifications in modular (#34866)

yonigozlan committed 1 year ago

Verified 3a8eb746

Bitnet test fix to avoid using gated model (#34863)

MekkCyber committed 1 year ago

Verified 54be2d7a

[CI] Skip EETQ tests while package is broken with latest transformers (#34854)

BenjaminBossan committed 1 year ago

Verified 286ffaaf

smol improvements to support more flexible usage (#34857)

andimarafioti committed 1 year ago

Verified 861758e2

Speculative decoding: Test the target distribution (to prevent issues like #32867) (#34553)

Nadav Timor committed 1 year ago

Verified 42b36d73

Auto compile when static cache (#34247)

ArthurZucker committed 1 year ago

Verified 597efd21

Remove quantization related config from dequantized model (#34856)

konradkalita committed 1 year ago

Verified d9e6f307

Update checks for torch.distributed.tensor to require torch >= 2.5 (#34816)

loadams committed 1 year ago

Verified 1867be66

Watermarking: fix order (#34849)

zucchini-nlp committed 1 year ago

Verified 6a912ff2

Refactor StarCoder2 using modular (#34015)

Cyrilvallez committed 1 year ago

Verified 4e90b99e

Fix heuristic scheduling for UAG (#34805)

jmamou committed 1 year ago

Verified 18871599

Fix ds nvme (#34444)

eljandoubi committed 1 year ago

Verified d6a5c23f

Improve gguf tensor processing (#34515)

VladOS95-cyber committed 1 year ago

Verified ae5cbf80

Add Nemotron GGUF Loading Support (#34725)

farrosalferro committed 1 year ago

Verified c57eafda

Change logging level from warning to info for `max_steps` overriding `num_train_epochs` (#34810)

qgallouedec committed 1 year ago

Verified d4e1acbb

VLMs: enable generation tests - last batch (#34484)

zucchini-nlp committed 1 year ago

Verified 28fb02fc

Fix CI slack reporting issue (#34833)

ydshieh committed 1 year ago

Verified 40821a24

Fix CI by tweaking torchao tests (#34832)

SunMarc committed 1 year ago

Verified 3cb8676a

Fix hyperparameter search when optuna+deepseed (#34642)

corentin-ryr committed 1 year ago

Verified bf42c3bd

Torchao weights only + prequantized compability (#34355)

SunMarc committed 1 year ago

Verified 67890de3

Fix: take into account meta device (#34134)

tibor-reiss committed 1 year ago

Verified f297af55

Newer Older