Source code for deeptab.models.mlp

from deeptab.architectures.mlp import MLP
from deeptab.models.classifier_base import SklearnBaseClassifier
from deeptab.models.lss_base import SklearnBaseLSS
from deeptab.models.regressor_base import SklearnBaseRegressor

from ..configs.models.mlp_config import MLPConfig
from ._docstring import generate_docstring




class MLPRegressor(SklearnBaseRegressor):
    _model_cls = MLP
    _config_cls = MLPConfig

    __doc__ = generate_docstring(
        MLPConfig,
        model_description="""
        Multi-Layer Perceptron regressor. This class extends the SklearnBaseRegressor class and uses the MLP model
        with the default MLP configuration.
        """,
        examples="""
        >>> from deeptab.models import MLPRegressor
        >>> from deeptab.configs import MLPConfig, TrainerConfig
        >>> model = MLPRegressor(
        ...     model_config=MLPConfig(layer_sizes=[128, 64]),
        ...     trainer_config=TrainerConfig(max_epochs=100, lr=1e-3),
        ... )
        >>> model.fit(X_train, y_train)
        >>> preds = model.predict(X_test)
        """,
    )





class MLPClassifier(SklearnBaseClassifier):
    _model_cls = MLP
    _config_cls = MLPConfig

    __doc__ = generate_docstring(
        MLPConfig,
        model_description="""
        Multi-Layer Perceptron classifier This class extends the SklearnBaseClassifier class and uses the MLP model
        with the default MLP configuration.
        """,
        examples="""
        >>> from deeptab.models import MLPClassifier
        >>> from deeptab.configs import MLPConfig, TrainerConfig
        >>> model = MLPClassifier(
        ...     model_config=MLPConfig(layer_sizes=[128, 64]),
        ...     trainer_config=TrainerConfig(max_epochs=100, lr=1e-3),
        ... )
        >>> model.fit(X_train, y_train)
        >>> preds = model.predict(X_test)
        """,
    )





class MLPLSS(SklearnBaseLSS):
    _model_cls = MLP
    _config_cls = MLPConfig

    __doc__ = generate_docstring(
        MLPConfig,
        model_description="""
        Multi-Layer Perceptron for distributional regression. This class extends the SklearnBaseLSS class and uses the MLP model
        with the default MLP configuration.
        """,
        examples="""
        >>> from deeptab.models import MLPLSS
        >>> from deeptab.configs import MLPConfig
        >>> model = MLPLSS(model_config=MLPConfig(layer_sizes=[128, 64]))
        >>> model.fit(X_train, y_train, family='normal')
        >>> preds = model.predict(X_test)
        >>> model.evaluate(X_test, y_test)
        """,
    )