nemo_automodel.components.launcher.nemo_run.config

Module Contents

Classes

Name	Description
`NemoRunConfig`	Configuration for the NeMo-Run launcher backend.

Data

DEFAULT_EXECUTORS_FILE

_LAUNCHER_KEYS

_NEMORUN_HOME

API

class nemo_automodel.components.launcher.nemo_run.config.NemoRunConfig(
    executor: str = 'local',
    job_name: str = '',
    detach: bool = True,
    tail_logs: bool = False,
    executors_file: str = (lambda: DEFAULT_EXECUTORS_...,
    job_dir: str = '',
    overrides: dict = dict()
)

Dataclass

Configuration for the NeMo-Run launcher backend.

The executor field selects a named executor from $NEMORUN_HOME/executors.py, or "local" for local execution.

Any key not recognised as a launcher setting is collected into overrides and applied directly to the executor via setattr. This means any executor attribute (nodes, partition, container_image, time, env_vars, etc.) can be overridden from YAML without changes to this config class.

detach

bool = True

executor

str = 'local'

executors_file

str

job_dir

str = ''

job_name

str = ''

overrides

dict = field(default_factory=dict)

tail_logs

bool = False

nemo_automodel.components.launcher.nemo_run.config.NemoRunConfig.from_dict(
    d: dict
) -> 'NemoRunConfig'

classmethod

Build from a raw YAML dict, splitting launcher keys from executor overrides.

nemo_automodel.components.launcher.nemo_run.config.DEFAULT_EXECUTORS_FILE = os.path.join(_NEMORUN_HOME, 'executors.py')

nemo_automodel.components.launcher.nemo_run.config._LAUNCHER_KEYS = frozenset({'executor', 'job_name', 'detach', 'tail_logs', 'executors_file', 'job...

nemo_automodel.components.launcher.nemo_run.config._NEMORUN_HOME = os.environ.get('NEMORUN_HOME', os.path.join(os.path.expanduser('~'), '.nemo_run'...