smash.factory.Net

class smash.factory.Net

Artificial Neural Network initialization.

Attributes:

history

A dictionary saving training information.

layers

List of Layer objects defining the graph of the network.

Methods

add_conv2d(filters, filter_shape[, ...])	Add a 2D convolutional layer (with same padding and a stride of one) to the neural network.
add_dense(neurons[, input_shape, ...])	Add a fully-connected layer to the neural network.
add_dropout(drop_rate)	Add a dropout layer that randomly sets the output of the previous layer to zero with a specified probability.
add_flatten()	Add a flatten layer to reshape the input from 2D layer into 1D layer.
add_scale(bounds)	Add a scaling layer that applies the min-max scaling function to the outputs.
copy()	Make a deepcopy of the Net.
forward_pass(x)	Perform a forward pass through the neural network.
get_bias()	Get the biases of the trainable layers of the neural network Net.
get_weight()	Get the weights of the trainable layers of the neural network Net.
set_bias([value, random_state])	Set the values of the bias in the neural network Net.
set_trainable(trainable)	Method which enables to train or freeze the weights and biases of the network's layers.
set_weight([value, random_state])	Set the values of the weight in the neural network Net.

Examples

>>> from smash.factory import Net
>>> net = Net()
>>> net
+------------------------------------------------+
| Layer Type  Input/Output Shape  Num Parameters |
+------------------------------------------------+
Total parameters: 0
Trainable parameters: 0