Rémi Ouazan eef9fb2af3 Fix EncoderDecoder cache (#41612) ...

* Fix EncoderDecoder cache

* Add the option for the ddp data tuples to have 2 elems

* Modifiy the order of the KV and sliding

* Adapted RAG and Whisper to new EncoderDecoderCache

* A single comma

* Remove kwargs in map

* Fixed order in manual injection cache test

* Slight changes to support legacy format

* Removed Nonnes

2025-10-16 14:55:41 +02:00

..

cli

Migrate transformers cli to Typer (#41487)

2025-10-16 13:29:42 +02:00

deepspeed

Fix trainer simple tests (#41449)

2025-10-15 14:09:00 +02:00

extended

deprecate overwrite_output_dir (#41323)

2025-10-09 18:36:19 +02:00

fixtures

Add Parakeet (#39062)

2025-09-25 13:52:24 +00:00

fsdp

deprecate overwrite_output_dir (#41323)

2025-10-09 18:36:19 +02:00

generation

Fix some tests (#41503)

2025-10-10 11:05:09 +02:00

models

[Executorch] Simplify for encoder models (#41627)

2025-10-16 13:57:52 +02:00

optimization

[tests] remove TF tests (uses of require_tf) (#38944)

2025-06-25 17:29:10 +00:00

peft_integration

FIX: Bug in PEFT integration delete_adapter method (#41252)

2025-10-02 18:36:57 +02:00

pipelines

enable new model uts to xpu and fix some failures on xpu (#41386)

2025-10-08 10:14:50 +00:00

quantization

Fix FP-Quant quantization fallback CPU dispatch. (#41619)

2025-10-16 11:41:01 +00:00

repo_utils

Standardize PretrainedConfig to PreTrainedConfig (#41300)

2025-10-06 11:34:02 +02:00

sagemaker

Remove local_rank arg from TrainingArguments (#41382)

2025-10-09 18:54:12 +02:00

tensor_parallel

Allow passing tp_plan in from_pretrained directly (#41435)

2025-10-16 11:12:07 +02:00

tokenization

delete some tokenizer tests using pickle (#41514)

2025-10-14 14:50:51 +02:00

trainer

Fix trainer simple tests (#41449)

2025-10-15 14:09:00 +02:00

utils

Fix EncoderDecoder cache (#41612)

2025-10-16 14:55:41 +02:00

__init__.py

GPU text generation: mMoved the encoded_prompt to correct device

2020-01-06 15:11:12 +01:00

causal_lm_tester.py

[causallm tester] automate pipeline mappings + bloom tests (#41318)

2025-10-10 10:02:00 +01:00

test_backbone_common.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_configuration_common.py

Subconfig is a class attribute (#41308)

2025-10-09 10:46:44 +02:00

test_executorch.py

Support input_embeds in torch exportable decoders (#39836)

2025-08-07 08:51:31 +00:00

test_feature_extraction_common.py

Use Python 3.9 syntax in tests (#37343)

2025-04-08 14:12:08 +02:00

test_image_processing_common.py

Validate processing kwargs with @strict from huggingface_hub (#40793)

2025-10-08 16:14:09 +02:00

test_image_transforms.py

Raise TypeError instead of ValueError for invalid types (#38660)

2025-07-21 12:42:00 +00:00

test_modeling_common.py

Gemma3 fixes (#41572)

2025-10-14 18:33:27 +02:00

test_pipeline_mixin.py

🚨🚨🚨 Fully remove Tensorflow and Jax support library-wide (#40760)

2025-09-18 18:27:39 +02:00

test_processing_common.py

🚨 [v5] Toggle the serialization format in processors (#41474)

2025-10-16 10:19:22 +02:00

test_sequence_feature_extraction_common.py

[tests] remove TF tests (uses of require_tf) (#38944)

2025-06-25 17:29:10 +00:00

test_tokenization_common.py

Fix tokenization test (#41649)

2025-10-16 11:14:20 +02:00

test_tokenization_mistral_common.py

Patch MistralCommonTokenizer (#41439)

2025-10-14 11:13:19 +00:00

test_training_args.py

Fix TrainingArguments.torch_empty_cache_steps post_init check (#36734)

2025-03-17 16:09:46 +01:00

test_video_processing_common.py

Validate processing kwargs with @strict from huggingface_hub (#40793)

2025-10-08 16:14:09 +02:00