[Multimodal] Type dummy options per modality (#57576)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 5
parent
56cac80b14
commit
8a5cf54387
@@ -26,7 +26,7 @@ from transformers.feature_extraction_utils import BatchFeature
|
||||
from transformers.modeling_outputs import BaseModelOutput
|
||||
|
||||
from vllm.config import VllmConfig
|
||||
from vllm.config.multimodal import AudioDummyOptions, BaseDummyOptions
|
||||
from vllm.config.multimodal import MultiModalDummyOptions
|
||||
from vllm.inputs import MultiModalDataDict
|
||||
from vllm.model_executor.layers.attention.mm_encoder_attention import MMEncoderAttention
|
||||
from vllm.model_executor.layers.linear import QKVParallelLinear, RowParallelLinear
|
||||
@@ -580,7 +580,7 @@ class FunAudioChatDummyInputsBuilder(
|
||||
self,
|
||||
seq_len: int,
|
||||
mm_counts: Mapping[str, int],
|
||||
mm_options: Mapping[str, BaseDummyOptions],
|
||||
mm_options: MultiModalDummyOptions,
|
||||
) -> MultiModalDataDict:
|
||||
feature_extractor = self.info.get_feature_extractor()
|
||||
sampling_rate = int(feature_extractor.sampling_rate)
|
||||
@@ -597,15 +597,12 @@ class FunAudioChatDummyInputsBuilder(
|
||||
1,
|
||||
(target_num_frames * sampling_rate + token_fps - 1) // token_fps,
|
||||
)
|
||||
num_audios = int(mm_counts.get("audio", 0))
|
||||
|
||||
audio_overrides = mm_options.get("audio")
|
||||
assert audio_overrides is None or isinstance(audio_overrides, AudioDummyOptions)
|
||||
return {
|
||||
"audio": self._get_dummy_audios(
|
||||
length=audio_len,
|
||||
num_audios=num_audios,
|
||||
overrides=audio_overrides,
|
||||
num_audios=int(mm_counts.get("audio", 0)),
|
||||
overrides=mm_options.get("audio"),
|
||||
)
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user