cellink.tl.external.LIVIRunner

class cellink.tl.external.LIVIRunner(livi_root, execution_mode='python_api', python_executable='python', device='auto')

Manages access to the LIVI package.

Parameters:

• livi_root (str or Path) – Path to the LIVI repository root (must contain src/ and configs/ directories).

• execution_mode ({"python_api", "subprocess"}) – "python_api" (default) imports LIVI directly into the current process — no Hydra overhead, returns Python objects. "subprocess" runs LIVI’s src/train.py as a child process via the Hydra CLI, which is better suited for isolated HPC job submission.

• python_executable (str) – Python binary used in "subprocess" mode.

• device (str) – Compute device: "auto" (detect GPU), "cuda", or "cpu".

Methods table

get_association_testing_fn()	Import and return run_LIVI_genetic_association_testing.
get_datamodule_classes()	Import and return (LIVIDataModule, LIVIDataset).
get_livi_class()	Import and return the LIVI model class.
resolve_device()	Return the resolved device string ("cuda" or "cpu").

Methods

LIVIRunner.get_association_testing_fn()

Import and return run_LIVI_genetic_association_testing.

LIVIRunner.get_datamodule_classes()

Import and return (LIVIDataModule, LIVIDataset).

LIVIRunner.get_livi_class()

Import and return the LIVI model class.

LIVIRunner.resolve_device()

Return the resolved device string ("cuda" or "cpu").

Return type:

str