biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Misc] Automatically resolve HF processor init kwargs (#22005)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-08-01 13:44:10 +08:00
committed by GitHub
parent ad57f23f6a
commit 82de9b9d46
40 changed files with 334 additions and 727 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
vllm/model_executor/models/deepseek_vl2.py
View File

@@ -214,25 +214,25 @@ class DeepseekVL2MultiModalProcessor(
mm_kwargs: Mapping[str, object],
tok_kwargs: Mapping[str, object],
) -> BatchFeature:
if mm_data:
processed_outputs = self.info.ctx.call_hf_processor(
self.info.get_hf_processor(**mm_kwargs),
dict(prompt=prompt, **mm_data),
dict(**mm_kwargs, **tok_kwargs),
)
pixel_values = processed_outputs["pixel_values"]
# split pixel values into patches corresponding to each image
images_spatial_crop = processed_outputs["images_spatial_crop"]
patches_per_image = [
x.prod().item() + 1 for x in images_spatial_crop
]
pixel_values = pixel_values.split(patches_per_image)
processed_outputs["pixel_values"] = pixel_values
else:
if not mm_data:
tokenizer = self.info.get_tokenizer()
processed_outputs = tokenizer(prompt,
add_special_tokens=True,
return_tensors="pt")
return tokenizer(prompt,
add_special_tokens=True,
return_tensors="pt")
processed_outputs = super()._call_hf_processor(
prompt=prompt,
mm_data=mm_data,
mm_kwargs=mm_kwargs,
tok_kwargs=tok_kwargs,
)
pixel_values = processed_outputs["pixel_values"]
# split pixel values into patches corresponding to each image
images_spatial_crop = processed_outputs["images_spatial_crop"]
patches_per_image = [x.prod().item() + 1 for x in images_spatial_crop]
pixel_values = pixel_values.split(patches_per_image)
processed_outputs["pixel_values"] = pixel_values
return processed_outputs