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

Add code-revision config argument for Hugging Face Hub (#2892)

Browse Source
This commit is contained in:
Mark Mozolewski
2024-02-17 22:36:53 -08:00
committed by GitHub
parent 8f36444c4f
commit 786b7f18a5
3 changed files with 30 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
vllm/engine/arg_utils.py
View File

@@ -32,6 +32,7 @@ class EngineArgs:
max_paddings: int = 256
disable_log_stats: bool = False
revision: Optional[str] = None
code_revision: Optional[str] = None
tokenizer_revision: Optional[str] = None
quantization: Optional[str] = None
enforce_eager: bool = False
@@ -75,6 +76,13 @@ class EngineArgs:
help='the specific model version to use. It can be a branch '
'name, a tag name, or a commit id. If unspecified, will use '
'the default version.')
parser.add_argument(
'--code-revision',
type=str,
default=None,
help='the specific revision to use for the model code on '
'Hugging Face Hub. It can be a branch name, a tag name, or a '
'commit id. If unspecified, will use the default version.')
parser.add_argument(
'--tokenizer-revision',
type=str,
@@ -279,13 +287,12 @@ class EngineArgs:
) -> Tuple[ModelConfig, CacheConfig, ParallelConfig, SchedulerConfig,
DeviceConfig, Optional[LoRAConfig]]:
device_config = DeviceConfig(self.device)
model_config = ModelConfig(self.model, self.tokenizer,
self.tokenizer_mode, self.trust_remote_code,
self.download_dir, self.load_format,
self.dtype, self.seed, self.revision,
self.tokenizer_revision, self.max_model_len,
self.quantization, self.enforce_eager,
self.max_context_len_to_capture)
model_config = ModelConfig(
self.model, self.tokenizer, self.tokenizer_mode,
self.trust_remote_code, self.download_dir, self.load_format,
self.dtype, self.seed, self.revision, self.code_revision,
self.tokenizer_revision, self.max_model_len, self.quantization,
self.enforce_eager, self.max_context_len_to_capture)
cache_config = CacheConfig(self.block_size,
self.gpu_memory_utilization,
self.swap_space, self.kv_cache_dtype,