ml_grid.model_classes.keras_classifier_class

Classes

KerasClassifierClass

Initializes the KerasClassifierClass.

Functions

create_model(→ keras.models.Sequential)

Builds and compiles a Keras Sequential model.

Module Contents

ml_grid.model_classes.keras_classifier_class.create_model(layers: int = 1, l1_reg: float = 0.0, l2_reg: float = 0.0, width: int = 15, learning_rate: float = 0.01, dropout_val: float = 0.2, input_dim_val: int = 0) → keras.models.Sequential

Builds and compiles a Keras Sequential model.

Parameters:

• layers (int) – The number of dense layers in the model.

• l1_reg (float) – L1 regularization factor.

• l2_reg (float) – L2 regularization factor.

• width (int) – The number of units in each dense layer.

• learning_rate (float) – The learning rate for the Adam optimizer.

• dropout_val (float) – The dropout rate.

• input_dim_val (int) – The input dimension for the first layer.

Returns:

The compiled Keras model.

Return type:

Sequential

class ml_grid.model_classes.keras_classifier_class.KerasClassifierClass(X: pandas.DataFrame, y: pandas.Series, parameter_space_size: str | None = None)

Initializes the KerasClassifierClass.

This configures a Keras Sequential model for binary classification, wrapped in a KerasClassifier to be compatible with scikit-learn’s hyperparameter tuning utilities.

Parameters:

• X (pd.DataFrame) – Feature matrix for training.

• y (pd.Series) – Target vector for training.

• parameter_space_size (Optional[str]) – Size of the parameter space for optimization. Defaults to None.

Raises:

ValueError – If parameter_space_size is not a valid key (though current implementation does not explicitly raise this).

X: pandas.DataFrame

y: pandas.Series

x_train_col_val: int

method_name: str = 'KerasClassifier'

parameter_space: Dict[str, Any]

algorithm_implementation