[model] support audio (#6701)

* support qwen2_audio

* improve code

* lint

* fix

* fix

* fix

---------

Co-authored-by: hiyouga <hiyouga@buaa.edu.cn>
Former-commit-id: 24c7842948
This commit is contained in:
Zhangchi Feng
2025-02-05 04:59:09 +08:00
committed by GitHub
parent e665e1fed5
commit 01915eaf40
37 changed files with 736 additions and 213 deletions

View File

@@ -49,6 +49,7 @@ class PairwiseDataCollatorWithPadding(MultiModalDataCollatorForSeq2Seq):
"labels": feature["chosen_input_ids"] if self.train_on_prompt else feature["chosen_labels"],
"images": feature["images"],
"videos": feature["videos"],
"audios": feature["audios"],
}
)