bgd.layers.fc module

class bgd.layers.fc.FullyConnected(n_in, n_out, copy=False, with_bias=True, dtype=<class 'numpy.float32'>, initializer=<bgd.initializers.GlorotUniformInitializer object>, bias_initializer=<bgd.initializers.ZeroInitializer object>)

Bases: bgd.layers.layer.Layer

Fully connected (dense) neural layer. Each output neuron is a weighted sum of its inputs with possibly a bias.

Parameters

• n_in (int) – Number of input neurons.

• n_out (int) – Number of output neurons.

• copy (bool) – Whether to copy layer output.

• with_bias (bool) – Whether add a bias to output neurons.

• dtype (type) – Type of weights and biases.

• initializer (bgd.initializers.Initializer) – Initializer of the weights.

• bias_initializer (bgd.initializers.Initializer) – Initializer of the biases.

weights

Matrix of weights.

Type

np.ndarray

biases

Vector of biases.

Type

np.ndarray

_backward(error)

Wrapped method for applying a backward pass on input X.

_forward(X)

Wrapped method for applying a forward pass on input X.

get_parameters()

Retuns a tuple containing the parameters of the layer. If layer is non-learnable, None is returned instead. The tuple can have a size > 1 for convenience. For example, a convolutional neural layer has one array for the filters weights and one array for the biases.

update_parameters(delta_fragments)

Update parameters: base class has no learnable parameters. Subclasses that are learnable must override this method.

Parameters

delta_fragments (tuple) – Tuple of NumPy arrays. Each array is a parameter update vector and is used to update parameters using the following formula: params = params - delta_fragment. The tuple can have a size > 1 for convenience. For example, a convolutional neural layer has one array to update the filters weights and one array to update the biases: weights = weights - delta_fragments[0] biases = biases - delta_fragments[1]