Skip to main content

mellea.formatters.granite.intrinsics.util

Common utility functions for this package.

Functions​

FUNC make_config_dict ​

make_config_dict(config_file: str | pathlib.Path | None = None, config_dict: dict | None = None) -> dict | None

Create a configuration dictionary from YAML file or dict.

This function is not a public API and is not intended for use outside this library.

Common initialization code for reading YAML config files in factory classes. Also parses JSON fields.

Args:

  • config_file: Path to a YAML configuration file. Exactly one of config_file and config_dict must be provided.
  • config_dict: Pre-parsed configuration dict (from yaml.safe_load()). Exactly one of config_file and config_dict must be provided.

Returns:

  • Validated configuration dict with optional fields set to None and JSON
  • string fields parsed to Python objects.

Raises:

  • ValueError: If both or neither of config_file and config_dict are provided, if a required field is missing, if an unexpected top-level field is encountered, or if a JSON field cannot be parsed.

FUNC adapter_subpath ​

adapter_subpath(alora: bool = False) -> str

Return the Hugging Face Hub subpath where an intrinsic's adapter lives.

Encapsulates the layout convention used by the Granite Intrinsics Library and related repositories so callers don't replicate the rules. Both :func:obtain_lora and out-of-tree consumers (e.g. drift checks in tests) should call this function rather than building the path themselves.

Args:

  • intrinsic_name: Short name of the intrinsic model, such as "certainty".
  • target_model_name: Name of the base model for the LoRA or aLoRA adapter. May be a raw HF repo ID; canonical normalization is applied.
  • repo_id: Hugging Face Hub repository containing the adapter collection. Used to select between old and new directory layouts.
  • alora: If True, return the path for the aLoRA variant; otherwise LoRA.

Returns:

  • Subpath relative to the repo root, e.g. "certainty/granite-4.1-3b/lora".

FUNC obtain_lora ​

obtain_lora(revision: str = 'main', alora: bool = False, cache_dir: str | None = None, file_glob: str = '*') -> pathlib.Path

Download and cache an adapter that implements and intrinsic.

Downloads a LoRA or aLoRA adapter from a collection of adapters that follow the same layout as the Granite Intrinsics Library. Caches the downloaded adapter files on local disk.

Args:

  • intrinsic_name: Short name of the intrinsic model, such as "certainty".
  • target_model_name: Name of the base model for the LoRA or aLoRA adapter.
  • repo_id: Hugging Face Hub repository containing a collection of LoRA and/or aLoRA adapters for intrinsics.
  • revision: Git revision of the repository to download from.
  • alora: If True, load the aLoRA version of the intrinsic; otherwise use LoRA.
  • cache_dir: Local directory to use as a cache (Hugging Face Hub format), or None to use the default location.
  • file_glob: Only files matching this glob will be downloaded to the cache.

Returns:

  • Full path to the local copy of the specified (a)LoRA adapter, suitable for
  • passing to commands that serve the adapter.

Raises:

  • ValueError: If the specified intrinsic adapter cannot be found in the Hugging Face Hub repository at the expected path.

FUNC obtain_io_yaml ​

obtain_io_yaml(revision: str = 'main', alora: bool = False, cache_dir: str | None = None) -> pathlib.Path

Download cached io.yaml configuration file for an intrinsic.

Downloads an io.yaml configuration file for an intrinsic with a model repository that follows the format of the Granite Intrinsics Library if one is not already in the local cache.

Args:

  • intrinsic_name: Short name of the intrinsic model, such as "certainty".
  • target_model_name: Name of the base model for the LoRA or aLoRA adapter.
  • repo_id: Hugging Face Hub repository containing a collection of LoRA and/or aLoRA adapters for intrinsics.
  • revision: Git revision of the repository to download from.
  • alora: If True, load the aLoRA version of the intrinsic; otherwise use LoRA.
  • cache_dir: Local directory to use as a cache (Hugging Face Hub format), or None to use the default location.

Returns:

  • Full path to the local copy of the io.yaml file, suitable for passing to
  • IntrinsicsRewriter.