[MM] Remove text components from ProcessorInputs (#53093)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2026-08-20 20:42:53 -07:00
committed by GitHub
parent 2adf4b9e2a
commit b00f475f09
92 changed files with 220 additions and 806 deletions
+1 -11
View File
@@ -617,10 +617,9 @@ class FunAudioChatMultiModalProcessor(
prompt: str,
mm_data: Mapping[str, object],
mm_kwargs: Mapping[str, object],
tok_kwargs: Mapping[str, object],
) -> BatchFeature:
tokenizer = self.info.get_tokenizer()
input_ids = torch.tensor([tokenizer.encode(prompt, **tok_kwargs)])
input_ids = torch.tensor([tokenizer.encode(prompt)])
audios = mm_data.get("audios", [])
if not audios:
@@ -679,15 +678,6 @@ class FunAudioChatMultiModalProcessor(
return BatchFeature({"input_ids": input_ids, **mm_inputs})
def _hf_processor_applies_updates(
self,
prompt_text: str,
mm_items: MultiModalDataItems,
hf_processor_mm_kwargs: Mapping[str, object],
tokenization_kwargs: Mapping[str, object],
) -> bool:
return False
def _get_mm_fields_config(
self,
hf_inputs: BatchFeature,