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

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

Browse Source 
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
parent 1d4cfe2be1
commit e893795443
44 changed files with 249 additions and 579 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
tests/worker/test_encoder_decoder_model_runner.py
View File

@@ -19,14 +19,7 @@ def _create_model_runner(model: str, *args,
engine_args = EngineArgs(model, *args, **kwargs)
engine_config = engine_args.create_engine_config()
model_runner = EncoderDecoderModelRunner(
model_config=engine_config.model_config,
parallel_config=engine_config.parallel_config,
scheduler_config=engine_config.scheduler_config,
device_config=engine_config.device_config,
cache_config=engine_config.cache_config,
load_config=engine_config.load_config,
lora_config=engine_config.lora_config,
prompt_adapter_config=engine_config.prompt_adapter_config,
vllm_config=engine_config,
is_driver_worker=True,
)
return model_runner