sega_learn.neural_networks.neuralNetworkBase

Modules
		cupy numpy numba.core.types

Classes
		builtins.object NeuralNetworkBase   class NeuralNetworkBase(builtins.object)     NeuralNetworkBase(layers, dropout_rate=0.0, reg_lambda=0.0, activations=None, loss_function=None, regressor=False)   NeuralNetworkBase is an abstract base class for building neural networks.   It provides a framework for initializing layers, performing forward and backward propagation, training, evaluating, and predicting with a neural network. Subclasses should implement the abstract methods to define specific behavior.   Attributes:     layer_sizes (list): Sizes of the layers in the network.     dropout_rate (float): Dropout rate for regularization.     reg_lambda (float): Regularization strength for L2 regularization.     activations (list): Activation functions for each layer.     layers (list): List of layer objects or configurations.     weights (list): List of weight matrices for the layers.     biases (list): List of bias vectors for the layers.     layer_outputs (ndarray): Outputs of each layer during forward propagation.     is_binary (bool): Whether the network is for binary classification.   Methods:     __init__(layers, dropout_rate=0.0, reg_lambda=0.0, activations=None, loss_function=None, regressor=False):         Initializes the neural network with the given layers and parameters.     initialize_layers():         Abstract method to initialize the weights and biases of the layers.     forward(X, training=True):         Abstract method to perform forward propagation through the network.     backward(y):         Abstract method to perform backward propagation through the network.     train(X_train, y_train, X_val=None, y_val=None, optimizer=None, epochs=100,             batch_size=32, early_stopping_threshold=10, lr_scheduler=None, p=True,             use_tqdm=True, n_jobs=1, track_metrics=False, track_adv_metrics=False):         Abstract method to train the neural network using the provided training data.     evaluate(X, y):         Abstract method to evaluate the neural network on the provided data.     predict(X):         Abstract method to make predictions using the trained neural network.     calculate_loss(X, y):         Abstract method to calculate the loss of the neural network.     apply_dropout(X):         Applies dropout to the activation values for regularization.     compute_l2_reg(weights):         Computes the L2 regularization term for the given weights.     calculate_precision_recall_f1(X, y):         Calculates precision, recall, and F1 score for the predictions.     create_scheduler(scheduler_type, optimizer, **kwargs):         Creates a learning rate scheduler based on the specified type.     plot_metrics(save_dir=None):         Plots the training and validation metrics, including loss, accuracy,         learning rate, and optionally precision, recall, and F1 score.     Methods defined here: __init__(self, layers, dropout_rate=0.0, reg_lambda=0.0, activations=None, loss_function=None, regressor=False) Initializes the neural network with the specified layers, dropout rate, regularization, and activations.   Args:     layers: (list) - A list of integers representing the sizes of each layer or a list of Layer objects.     dropout_rate: (float), optional - The dropout rate for regularization (default is 0.0).     reg_lambda: (float), optional - The regularization strength (default is 0.0).     activations: (list of str), optional - A list of activation functions for each layer (default is None, which sets "relu" for hidden layers and "softmax" for the output layer).     loss_function: (callable), optional - Custom loss function to use (default is None, which uses the default calculate_loss implementation).     regressor: (bool), optional - If True, the network is treated as a regressor (default is False).   Raises:     ValueError: If `layers` is not a list of integers or a list of Layer objects. apply_dropout(self, X) Applies dropout to the activation X.   Args:     X: (ndarray) - Activation values.   Returns:     ndarray: Activation values after applying dropout. backward(self, y) Performs backward propagation through the network. calculate_loss(self, X, y) Calculates the loss of the neural network. calculate_precision_recall_f1(self, X, y) Calculates precision, recall, and F1 score.   Args:     X: (ndarray) - Input data     y: (ndarray) - Target labels Returns:     precision: (float) - Precision score     recall: (float) - Recall score     f1: (float) - F1 score compute_l2_reg(self, weights) Computes the L2 regularization term.   Args:     weights: (list) - List of weight matrices.   Returns:     float: L2 regularization term. create_scheduler(self, scheduler_type, optimizer, **kwargs) Creates a learning rate scheduler. evaluate(self, X, y) Evaluates the neural network on the provided data. forward(self, X, training=True) Performs forward propagation through the network. initialize_layers(self) Initializes the weights and biases of the layers. plot_metrics(self, save_dir=None) Plots the training and validation metrics. predict(self, X) Makes predictions using the trained neural network. train(self, X_train, y_train, X_val=None, y_val=None, optimizer=None, epochs=100, batch_size=32, early_stopping_threshold=10, lr_scheduler=None, p=True, use_tqdm=True, n_jobs=1, track_metrics=False, track_adv_metrics=False) Trains the neural network using the provided training data. Data descriptors defined here: __dict__ dictionary for instance variables __weakref__ list of weak references to the object

Data
		CACHE = False conv_spec = [('in_channels', int32), ('out_channels', int32), ('kernel_size', int32), ('stride', int32), ('padding', int32), ('weights', Array(float64, 4, 'A', False, aligned=True)), ('biases', Array(float64, 2, 'A', False, aligned=True)), ('activation', unicode_type), ('weight_gradients', Array(float64, 4, 'A', False, aligned=True)), ('bias_gradients', Array(float64, 2, 'A', False, aligned=True)), ('input_cache', Array(float64, 4, 'A', False, aligned=True)), ('X_cols', Array(float64, 3, 'A', False, aligned=True)), ('X_padded', Array(float64, 4, 'A', False, aligned=True)), ('h_out', int32), ('w_out', int32), ('input_size', int32), ('output_size', int32)] flatten_spec = [('input_shape', UniTuple(int32, 3)), ('output_size', int32), ('input_cache', Array(float64, 4, 'A', False, aligned=True)), ('input_size', int32)] float64 = float64 int32 = int32 spec = [('weights', Array(float64, 2, 'C', False, aligned=True)), ('biases', Array(float64, 2, 'C', False, aligned=True)), ('activation', unicode_type), ('weight_gradients', Array(float64, 2, 'C', False, aligned=True)), ('bias_gradients', Array(float64, 2, 'C', False, aligned=True)), ('input_cache', Array(float64, 2, 'C', False, aligned=True)), ('output_cache', Array(float64, 2, 'C', False, aligned=True)), ('input_size', int32), ('output_size', int32)]