[Chore] Remove Sampler from Model Code (#17084)
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
@@ -9,7 +9,6 @@ from vllm.config import VllmConfig
|
||||
from vllm.logger import init_logger
|
||||
from vllm.model_executor.layers.layernorm import RMSNorm
|
||||
from vllm.model_executor.layers.logits_processor import LogitsProcessor
|
||||
from vllm.model_executor.layers.sampler import SamplerOutput
|
||||
from vllm.model_executor.layers.vocab_parallel_embedding import (
|
||||
DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead)
|
||||
from vllm.model_executor.model_loader.weight_utils import default_weight_loader
|
||||
@@ -131,10 +130,6 @@ class EAGLE(nn.Module):
|
||||
# checkpoint file has token_map tensor.
|
||||
self.token_map = None
|
||||
|
||||
@property
|
||||
def sampler(self):
|
||||
return self.model.sampler
|
||||
|
||||
def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor:
|
||||
return self.model.model.get_input_embeddings(input_ids)
|
||||
|
||||
@@ -188,14 +183,6 @@ class EAGLE(nn.Module):
|
||||
|
||||
return logits
|
||||
|
||||
def sample(
|
||||
self,
|
||||
logits: torch.Tensor,
|
||||
sampling_metadata: SamplingMetadata,
|
||||
) -> Optional[SamplerOutput]:
|
||||
next_tokens = self.sampler(logits, sampling_metadata)
|
||||
return next_tokens
|
||||
|
||||
def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
|
||||
# This implementation is incompitable with https://huggingface.co/yuhuili/EAGLE-LLaMA3-Instruct-8B
|
||||
# due to missing lm_head weights and its config being that of a
|
||||
|
||||
Reference in New Issue
Block a user