NielsRogge/transformers

Pull Requests Commits

Replace tokenizer=image_processor everywhere

NielsRogge committed 2 years ago

7c71afd9

Add image processor to trainer

NielsRogge committed 2 years ago

c5fb77c3

Reimplement "Automatic safetensors conversion when lacking these files" (#29846)

LysandreJik committed 2 years ago

Verified 4d8427f7

Fix 29807, sinusoidal positional encodings overwritten by post_init() (#29813)

hovnatan committed 2 years ago

Verified a81cf9ee

Mamba `slow_forward` gradient fix (#29563)

vasqu committed 2 years ago

Verified cefb819f

Add Qwen2MoE (#29377)

bozheng-hit committed 2 years ago

Verified 1c39974a

Support `num_attention_heads` != `num_key_value_heads` in Flax Llama Implementation (#29557)

bminixhofer committed 2 years ago

Verified 8e08acad

Set custom_container in build docs workflows (#29855)

Wauplin committed 2 years ago

Verified f01e1609

Disable AMD memory benchmarks (#29871)

IlyasMoutawwakil committed 2 years ago

Verified 07d79520

Add `cosine_with_min_lr` scheduler in Trainer (#29341)

liuyanyi committed 2 years ago

Verified ef609958

Allow `bos_token_id is None` during the generation with `inputs_embeds` (#29772)

LZHgrla committed 2 years ago

Verified 998b5bb5

[docs] Indent ordered list in add_new_model.md (#29796)

windsonsea committed 2 years ago

Verified b9ceb03d

Fix header in IFE task guide (#29859)

merveenoyan committed 2 years ago

Verified de81a677

Replace 'decord' with 'av' in VideoClassificationPipeline (#29747)

Tyx-main committed 2 years ago

Verified b32bf85b

Add warnings if training args differ from checkpoint trainer state (#29255)

jonflynng committed 2 years ago

Verified b5a6d6ee

remove quotes in code example (#29812)

johko committed 2 years ago

Verified 7eb3ba82

[`revert commit`] revert 00a09ed448082da3d6d35fb23a37b7d04f7b4dcd

ArthurZucker committed 2 years ago

e3e16ddc

ArthurZucker committed 2 years ago

00a09ed4

Populate torch_dtype from model to pipeline (#28940)

B-Step62 committed 2 years ago

Verified 8e9a2207

Fix the behavior of collecting 'num_input_tokens_seen' (#29099)

youliangh committed 2 years ago

Verified afe73aed

Remove static pretrained maps from the library's internals (#29112)

LysandreJik committed 2 years ago

Verified 39114c03

model_summary.md - Restore link to Harvard's Annotated Transformer. (#29702)

gamepad-coder committed 2 years ago

Verified 76a33a10

[DOCS] Fix typo for llava next docs (#29829)

aliencaocao committed 2 years ago

Verified dafe3702

[`SuperPoint`] Fix doc example (#29816)

amyeroberts committed 2 years ago

Verified c5f0288b

Complete security policy with mentions of remote code (#29707)

LysandreJik committed 2 years ago

Verified 7e1413d1

[`cleanup`] vestiges of causal mask (#29806)

ArthurZucker committed 2 years ago

Verified 2e7cb46f

replaced concatenation to f-strings to improve readability and unify … (#29785)

igeni committed 2 years ago

Verified 884b2215

Generate: remove unused attributes in `AssistedCandidateGenerator` (#29787)

gante committed 2 years ago

Verified 34e07f4b

rm input dtype change in CPU (#28631)

jiqing-feng committed 2 years ago

Verified e85654f5

Correct llava mask & fix missing setter for `vocab_size` (#29389)

fxmarty committed 2 years ago

Verified 13b23704

Older