huggingface/transformers

Pull Requests Commits

[Whisper Tokenizer] Encode timestamps (#26054)

sanchit-gandhi committed 2 years ago

Verified ac957f69

Fix eval accumulation when `accelerate` > 0.20.3 (#26060)

sam-scale committed 2 years ago

Verified 6d49b9dc

Add missing Maskformer dataclass decorator, add dataclass check in ModelOutput for subclasses (#25638)

rachthree committed 2 years ago

Verified d7bd325b

Flex xpu bug fix (#26135)

abhilash1910 committed 2 years ago

Verified 05de038f

[docs] last hidden state vs hidden_states[-1] (#26142)

MKhalusova committed 2 years ago

Verified 9709ab11

Update training_args.py - addition of self.distributed_state when using XPU (#25999)

Serizao committed 2 years ago

Verified e52f1cb6

Fix `beam_scores` shape when token scores shape changes after `logits_processor` (#25980)

BakerBunker committed 2 years ago

Verified 0fced067

Falcon: batched generation (#26137)

gante committed 2 years ago

Verified a796f7ee

Fix `test_finetune_bert2bert` (#25984)

ydshieh committed 2 years ago

Verified 95a90410

Generate: ignore warning when `generation_config.max_length` is set to `None` (#26147)

gante committed 2 years ago

Verified 86ffef87

docs: feat: add llama2 notebook resources from OSSCA community (#26076)

junejae committed 2 years ago

Verified a6ae2bd0

[`RWKV`] Final fix RWMV 4bit (#26134)

younesbelkada committed 2 years ago

Verified 7ccac73f

Update spectrogram and waveform model mapping for TTS/A pipeline (#26114)

Vaibhavs10 committed 2 years ago

Verified 32ec7345

Add missing space in generation/utils.py (#26121)

jbochi committed 2 years ago

Verified a9b63ca9

[`core`] fix 4bit `num_parameters` (#26132)

younesbelkada committed 2 years ago

Verified c8b26096

Fix AutoTokenizer docstring typo (#26117)

amyeroberts committed 2 years ago

Verified 7db1ad63

fix the deepspeed tests (#26021)

pacman100 committed 2 years ago

Verified b4773273

safeguard torch distributed check (#26056)

pacman100 committed 2 years ago

Verified 73b13ac0

Fix `MarianTokenizer` to remove metaspace character in `decode` (#26091)

tanaymeh committed 2 years ago

Verified 12f043ea

Text2text pipeline: don't parameterize from the config (#26118)

gante committed 2 years ago

Verified 03e309d5

chore: correct update_step and correct gradient_accumulation_steps (#26068)

pphuc25 committed 2 years ago

Verified 4fb64e28

enable optuna multi-objectives feature (#25969)

sywangyi committed 2 years ago

Verified 8f609ab9

🌐 [i18n-KO] Translated `contributing.md` to Korean (#25877)

mjk0618 committed 2 years ago

Verified 92f2fbad

[docs] Updates to TTS task guide with regards to the new TTS pipeline (#26095)

MKhalusova committed 2 years ago

Verified 1fe7ce48

🌐 [i18n-KO] Translated `llama2.md` to Korean (#26047)

mjk0618 committed 2 years ago

Verified be9438ed

Fix ExponentialDecayLengthPenalty negative logits issue (#25594)

pokjay committed 2 years ago

Verified 6acc27ee

Update logits_process.py docstrings (#25971)

larekrow committed 2 years ago

Verified d65c4a4f

Generate: legacy mode is only triggered when `generation_config` is untouched (#25962)

gante committed 2 years ago

Verified 3319eb54

[`core`] Import tensorflow inside relevant methods in `trainer_utils` (#26106)

younesbelkada committed 2 years ago

Verified 18abc756

[`Persimmon`] Add support for persimmon (#26042)

ArthurZucker committed 2 years ago

Verified 9cccb3a8

Newer Older