ertk.pytorch.models.PyTorchModelConfig

class ertk.pytorch.models.PyTorchModelConfig(optimiser: str = 'adam', opt_params: dict[str, typing.Any] = <factory>, learning_rate: float = 0.001, n_features: int = -1, n_classes: int = -1, loss: str = 'cross_entropy', loss_args: dict[str, typing.Any] = <factory>)

Bases: ERTKConfig

Base class for PyTorch model configuration dataclasses.

__init__(optimiser: str = 'adam', opt_params: dict[str, typing.Any] = <factory>, learning_rate: float = 0.001, n_features: int = -1, n_classes: int = -1, loss: str = 'cross_entropy', loss_args: dict[str, typing.Any] = <factory>) → None

Methods

__init__([optimiser, opt_params, ...])

Inherited Methods

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

Attributes

learning_rate	The learning rate to use.
loss	The loss function to use.
n_classes	The number of classes in the output data.
n_features	The number of features in the input data.
optimiser	The optimiser to use.
opt_params	Parameters to pass to the optimiser.
loss_args	Arguments to pass to the loss function.

learning_rate: float = 0.001

The learning rate to use.

loss: str = 'cross_entropy'

The loss function to use.

loss_args: dict[str, Any]

Arguments to pass to the loss function.

n_classes: int = -1

The number of classes in the output data.

n_features: int = -1

The number of features in the input data.

opt_params: dict[str, Any]

Parameters to pass to the optimiser.

optimiser: str = 'adam'

The optimiser to use.