[Misc] Consolidate ModelConfig code related to HF config (#10104)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2024-11-07 06:00:21 +00:00
committed by GitHub
parent 1fa020c539
commit db7db4aab9
10 changed files with 68 additions and 43 deletions
+1 -4
View File
@@ -151,7 +151,7 @@ class CPUWorker(LoraNotSupportedWorkerBase, LocalOrDistributedWorkerBase):
self.local_omp_cpuid = omp_cpuids.split("|")[rank]
ModelRunnerClass: Type[CPUModelRunner] = CPUModelRunner
if self._is_encoder_decoder_model():
if self.model_config.is_encoder_decoder:
ModelRunnerClass = CPUEncoderDecoderModelRunner
self.model_runner: CPUModelRunner = ModelRunnerClass(
vllm_config=vllm_config,
@@ -188,9 +188,6 @@ class CPUWorker(LoraNotSupportedWorkerBase, LocalOrDistributedWorkerBase):
raise RuntimeError("Profiler is not enabled.")
self.profiler.stop()
def _is_encoder_decoder_model(self):
return self.model_config.is_encoder_decoder_model
def init_device(self) -> None:
if self.local_omp_cpuid != "all":
ret = torch.ops._C_utils.init_cpu_threads_env(self.local_omp_cpuid)