Utility functions for vLLM config dataclasses.
 
   
   
  Source code in vllm/config/utils.py
   
  A decorator that ensures all fields in a dataclass have default values and that each field has a docstring.
If a ConfigT is used as a CLI argument itself, the type keyword argument provided by get_kwargs will be pydantic.TypeAdapter(ConfigT).validate_json(cli_arg) which treats the cli_arg as a JSON string which gets validated by pydantic.
Config validation is performed by the tools/pre_commit/validate_config.py script, which is invoked during the pre-commit checks.
Source code in vllm/config/utils.py
  
  Check if the text looks like a printed Python object, e.g. contains any substring matching the pattern: "at 0xFFFFFFF>" We match against 0x followed by 2-16 hex chars (there's a max of 16 on a 64-bit system).
Parameters:
| Name | Type | Description | Default | 
|---|---|---|---|
| text | str | The text to check | required | 
Returns:
| Name | Type | Description | 
|---|---|---|
| result | bool | 
 | 
Source code in vllm/config/utils.py
  
  Get any docstrings placed after attribute assignments in a class body.
https://davidism.com/mit-license/
Source code in vllm/config/utils.py
  
 get_field(cls: ConfigType, name: str) -> Field
Get the default factory field of a dataclass by name. Used for getting default factory fields in EngineArgs.
Source code in vllm/config/utils.py
  
  A helper function that retrieves an attribute from an object which may have multiple possible names. This is useful when fetching attributes from arbitrary transformers.PretrainedConfig instances.
Source code in vllm/config/utils.py
  
 is_init_field(cls: ConfigType, name: str) -> bool