ertk.train.EvalConfig

class ertk.train.EvalConfig(cv: CrossValidationConfig | None = None, train: DataSelector | None = None, valid: DataSelector | None = None, test: DataSelector | None = None, inner_kfold: int | None = None, inner_part: str | None = None)

Bases: ERTKConfig

Class to hold evaluation configuration.

__init__(cv: CrossValidationConfig | None = None, train: DataSelector | None = None, valid: DataSelector | None = None, test: DataSelector | None = None, inner_kfold: int | None = None, inner_part: str | None = None) → None

Methods

__init__([cv, train, valid, test, ...])

Inherited Methods

from_config(config)	Create config object from any compatible config.
from_file(path[, override])	Create config from YAML file and optionlly override some values.
merge_with_args([args])	Merge config with command-line arguments.
to_dictconfig()	Convert config to DictConfig.
to_file(path)	Write config to YAML file.
to_string()	Generate YAML string representation of config.

Attributes

cv	Cross-validation configuration.
inner_kfold	Number of folds for inner cross-validation, if using hyperparameter search.
inner_part	Partition to use for inner cross-validation, if using hyperparameter search.
test	Data selector for testing data.
train	Data selector for training data.
valid	Data selector for validation data.

cv: CrossValidationConfig | None = None

Cross-validation configuration. If not given, then no cross-validation is performed.

inner_kfold: int | None = None

Number of folds for inner cross-validation, if using hyperparameter search.

inner_part: str | None = None

Partition to use for inner cross-validation, if using hyperparameter search.

test: DataSelector | None = None

Data selector for testing data.

train: DataSelector | None = None

Data selector for training data.

valid: DataSelector | None = None

Data selector for validation data.