[2/N] executor pass the complete config to worker/modelrunner (#9938)

Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
This commit is contained in:
youkaichao
2024-11-02 07:35:05 -07:00
committed by GitHub
co-authored by Nick Hill
parent 1d4cfe2be1
commit e893795443
44 changed files with 249 additions and 579 deletions
+4 -22
View File
@@ -3,9 +3,7 @@ from typing import Any, Dict, List, Optional, Tuple, Type, Union
import torch
from vllm.config import (CacheConfig, DeviceConfig, LoadConfig, LoRAConfig,
ModelConfig, ObservabilityConfig, ParallelConfig,
PromptAdapterConfig, SchedulerConfig)
from vllm.config import VllmConfig
from vllm.distributed import get_pp_group
from vllm.forward_context import set_forward_context
from vllm.logger import init_logger
@@ -36,29 +34,13 @@ class EmbeddingModelRunner(
def __init__(
self,
model_config: ModelConfig,
parallel_config: ParallelConfig,
scheduler_config: SchedulerConfig,
device_config: DeviceConfig,
cache_config: CacheConfig,
load_config: LoadConfig,
lora_config: Optional[LoRAConfig],
vllm_config: VllmConfig,
kv_cache_dtype: Optional[str] = "auto",
is_driver_worker: bool = False,
prompt_adapter_config: Optional[PromptAdapterConfig] = None,
observability_config: Optional[ObservabilityConfig] = None,
):
super().__init__(model_config,
parallel_config,
scheduler_config,
device_config,
cache_config,
load_config,
lora_config=lora_config,
super().__init__(vllm_config=vllm_config,
kv_cache_dtype=kv_cache_dtype,
is_driver_worker=is_driver_worker,
prompt_adapter_config=prompt_adapter_config,
observability_config=observability_config)
is_driver_worker=is_driver_worker)
@torch.inference_mode()
def execute_model(