Source code for greenlearning.matrix_networks

from .neural_network import NeuralNetwork


[docs]def matrix_networks(layers, activation, shape): """Create a matrix of neural networks with the given parameters. Example: .. code-block:: python gl.matrix_networks([2] + [50] * 4 + [1], "rational", (2,1)) creates a matrix size 2 x 1 of rational networks with 4 hidden layers of 50 neurons. """ # Initialize the array M = [] # Create a matrix if len(shape) > 1: for i in range(shape[0]): Row = [] for j in range(shape[1]): Row.append(NeuralNetwork(layers, activation)) M.append(Row) # Create a vector else: for i in range(shape[0]): M.append(NeuralNetwork(layers, activation)) return M