kerasadf.layers.core.Dense

class kerasadf.layers.core.Dense(units, activation=None, use_bias=True, kernel_initializer='glorot_uniform', bias_initializer='zeros', kernel_regularizer=None, bias_regularizer=None, activity_regularizer=None, kernel_constraint=None, bias_constraint=None, **kwargs)

Bases: kerasadf.layers.core.ADFLayer

Densly-connected (fully connected) neural network layer.

Assumed Density Filtering (ADF) version of keras.layers.Dense.

Parameters

unitsint

Dimensionality of the output space (number of neurons).

activationcallable() or str, optional

Activation function to use. Default is no activation (ie. “linear” activation: a(x) = x).

use_biasbool

Whether the layer uses a bias vector.

kernel_initializerInitializer or str, optional

Initializer for the kernel weights matrix. Default is glorot_uniform initialization.

bias_initializerInitializer or str, optional

Initializer for the bias vector. Default is None.

kernel_regularizerRegularizer or str, optional

Regularizer function applied to the kernel weights matrix. Default is None.

bias_regularizerRegularizer or str, optional

Regularizer function applied to the bias vector. Default is None.

activity_regularizerRegularizer or str, optional

Regularizer function applied to the output of the layer. Default is None.

kernel_constraintConstraint or str, optional

Constraint function applied to the kernel weights matrix. Default is None.

bias_constraintConstraint or str, optional

Constraint function applied to the bias vector. Default is None.

Notes

Input shape

nD tensor with shape: (batch_size, ..., input_dim). The most common situation would be a 2D input with shape (batch_size, input_dim).

Output shape

nD tensor with shape: (batch_size, ..., units). For instance, for a 2D input with shape (batch_size, input_dim), the output would have shape (batch_size, units).

build(self, input_shape)

call(self, inputs)

compute_output_shape(self, input_shape)

get_config(self)