[V1] Change return type on get_multimodal_embeddings() (#19446)

Signed-off-by: Russell Bryant <rbryant@redhat.com>
This commit is contained in:
Russell Bryant
2025-06-16 13:32:15 -04:00
committed by GitHub
parent 387bdf0ab9
commit 90f9c2eb5c
37 changed files with 108 additions and 103 deletions
+4 -4
View File
@@ -478,11 +478,11 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal,
def get_language_model(self) -> torch.nn.Module:
return self.language_model
def get_multimodal_embeddings(
self, **kwargs: object) -> Optional[MultiModalEmbeddings]:
def get_multimodal_embeddings(self,
**kwargs: object) -> MultiModalEmbeddings:
image_input = self._parse_and_validate_image_input(**kwargs)
if image_input is None:
return None
return []
vision_embeddings = self._process_image_input(image_input)
return vision_embeddings
@@ -492,7 +492,7 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal,
multimodal_embeddings: Optional[MultiModalEmbeddings] = None,
) -> torch.Tensor:
if multimodal_embeddings is None:
if not multimodal_embeddings:
return self.language_model.get_input_embeddings(input_ids)
inputs_embeds = embed_multimodal(