[Chore] Remove Sampler from Model Code (#17084)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon
2025-04-24 02:49:33 -07:00
committed by GitHub
parent 2bc0f72ae5
commit b411418ff0
103 changed files with 48 additions and 1099 deletions
-10
View File
@@ -47,7 +47,6 @@ from vllm.model_executor.layers.linear import (ColumnParallelLinear,
RowParallelLinear)
from vllm.model_executor.layers.logits_processor import LogitsProcessor
from vllm.model_executor.layers.quantization import QuantizationConfig
from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
from vllm.model_executor.layers.vocab_parallel_embedding import (
DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead, VocabParallelEmbedding)
from vllm.model_executor.model_loader.weight_utils import (
@@ -1211,7 +1210,6 @@ class MllamaForConditionalGeneration(nn.Module, SupportsMultiModal,
)
self.logits_processor = LogitsProcessor(config.output_hidden_states,
config.text_config.vocab_size)
self.sampler = get_sampler()
def compute_logits(
self,
@@ -1222,14 +1220,6 @@ class MllamaForConditionalGeneration(nn.Module, SupportsMultiModal,
hidden_states, sampling_metadata)
return logits
def sample(
self,
logits: torch.Tensor,
sampling_metadata: SamplingMetadata,
) -> Optional[SamplerOutput]:
next_tokens = self.sampler(logits, sampling_metadata)
return next_tokens
def unpack_data(self,
image_data: Union[List[torch.Tensor], torch.Tensor],
padding_value=0) -> torch.Tensor: