Converter

class blackbox2c.Converter(config=None)

Main class that orchestrates the full conversion pipeline. Use this class when you need access to conversion metrics or want to reuse the surrogate tree.

Constructor

Parameter	Type	Default	Description
config	ConversionConfig	None	Conversion configuration. Uses defaults if None.

Methods

convert()

converter.convert(
    model, X_train,
    feature_names=None,
    class_names=None,
    X_test=None,
    target='c'
)

Run the full pipeline and return generated code.

Parameters: same as convert() convenience function, minus config and **config_kwargs.

Returns: str

get_metrics()

metrics = converter.get_metrics()

Returns a dict with conversion metrics:

{
    'fidelity': 0.974,              # agreement with original model
    'complexity': {
        'n_nodes': 45,
        'n_leaves': 23,
        'n_internal_nodes': 22,
        'max_depth': 5,
        'avg_path_length': 4.2,
    },
    'size_estimate': {
        'flash_bytes': 382,         # estimated FLASH usage
        'ram_bytes': 96,            # estimated RAM usage
    }
}

Attributes (after convert())

Attribute	Description
surrogate_tree_	The fitted DecisionTree (sklearn object)
feature_names_	Feature names used in code generation
class_names_	Class names (classification only)
metrics_	Same as get_metrics()

Example

from blackbox2c import Converter, ConversionConfig

config = ConversionConfig(max_depth=5, optimize_rules='high')
converter = Converter(config)

code = converter.convert(
    model, X_train,
    X_test=X_test,
    feature_names=['sl', 'sw', 'pl', 'pw'],
    class_names=['setosa', 'versicolor', 'virginica'],
    target='cpp',
)

m = converter.get_metrics()
print(f"Fidelity: {m['fidelity']:.3f}")
print(f"Nodes: {m['complexity']['n_nodes']}")