Configuration Utilities

API reference for configuration helpers that let you persist ANFIS setups, manage presets, and export trained models.

anfis_toolbox.config

Configuration utilities for ANFIS models.

ANFISConfig

ANFISConfig()

Configuration manager for ANFIS models.

__repr__

__repr__() -> str

String representation of configuration.

add_input_config

add_input_config(
    name: str,
    range_min: float,
    range_max: float,
    n_mfs: int = 3,
    mf_type: str = "gaussian",
    overlap: float = 0.5,
) -> ANFISConfig

Add input configuration.

Parameters:

Name	Type	Description	Default
name	str	Input variable name	required
range_min	float	Minimum input range	required
range_max	float	Maximum input range	required
n_mfs	int	Number of membership functions	3
mf_type	str	Type of membership functions	'gaussian'
overlap	float	Overlap factor	0.5

Returns:

Type	Description
ANFISConfig	Self for method chaining

build_model

build_model() -> TSKANFIS

Build ANFIS model from configuration.

Returns:

Type	Description
TSKANFIS	Configured ANFIS model

load classmethod

load(filepath: str | Path) -> ANFISConfig

Load configuration from JSON file.

Parameters:

Name	Type	Description	Default
filepath	str | Path	Path to configuration file	required

Returns:

Type	Description
ANFISConfig	ANFISConfig object

save

save(filepath: str | Path) -> None

Save configuration to JSON file.

Parameters:

Name	Type	Description	Default
filepath	str | Path	Path to save configuration file	required

set_training_config

set_training_config(
    method: str = "hybrid",
    epochs: int = 50,
    learning_rate: float = 0.01,
    verbose: bool = False,
) -> ANFISConfig

Set training configuration.

Parameters:

Name	Type	Description	Default
method	str	Training method ('hybrid' or 'backprop')	'hybrid'
epochs	int	Number of training epochs	50
learning_rate	float	Learning rate	0.01
verbose	bool	Whether to show training progress	False

Returns:

Type	Description
ANFISConfig	Self for method chaining

to_dict

to_dict() -> _ConfigDict

Convert configuration to dictionary.

Returns:

Type	Description
_ConfigDict	Configuration dictionary

ANFISModelManager

Model management utilities for saving/loading trained ANFIS models.

load_model staticmethod

load_model(filepath: str | Path) -> TSKANFIS

Load trained ANFIS model from file.

Parameters:

Name	Type	Description	Default
filepath	str | Path	Path to model file	required

Returns:

Type	Description
TSKANFIS	Loaded ANFIS model

save_model staticmethod

save_model(
    model: TSKANFIS,
    filepath: str | Path,
    include_config: bool = True,
) -> None

Save trained ANFIS model to file.

Parameters:

Name	Type	Description	Default
model	TSKANFIS	Trained ANFIS model	required
filepath	str | Path	Path to save model file	required
include_config	bool	Whether to save model configuration	True

create_config_from_preset

create_config_from_preset(preset_name: str) -> ANFISConfig

Create configuration from predefined preset.

Parameters:

Name	Type	Description	Default
preset_name	str	Name of predefined configuration	required

Returns:

Type	Description
ANFISConfig	ANFISConfig object

Raises:

Type	Description
ValueError	If preset name not found

list_presets

list_presets() -> dict[str, str]

List available predefined configurations.

Returns:

Type	Description
dict[str, str]	Dictionary mapping preset names to descriptions