Implements a sequential neural network. More...

Public Member Functions
procedure, pass, public	get_input_size => snn_get_input_size
	Returns the input size of the network. Implemented by snn_get_input_size. More...

procedure, pass, public	get_output_size => snn_get_output_size
	Returns the output size of the network. Implemented by snn_get_output_size. More...

procedure, pass, public	get_num_parameters => snn_get_num_parameters
	Returns the number of parameters. Implemented by snn_get_num_parameters. More...

procedure, pass, public	set_parameters => snn_set_parameters
	Setter for the network's parameters. Implemented by snn_set_parameters. More...

procedure, pass, public	get_parameters => snn_get_parameters
	Getter for the network's parameters. Implemented by snn_get_parameters. More...

procedure, pass, public	tofile => snn_tofile
	Saves the network. Implemented by snn_tofile. More...

procedure, pass, public	apply_forward => snn_apply_forward
	Applies and linearises the network. Implemented by snn_apply_forward. More...

procedure, pass, public	apply_tangent_linear => snn_apply_tangent_linear
	Applies the TL of the network. Implemented by snn_apply_tangent_linear. More...

procedure, pass, public	apply_adjoint => snn_apply_adjoint
	Applies the adjoint of the networ. Implemented by snn_apply_adjoint. More...

Public Attributes
integer(ik)	num_layers
	The number of layers. More...

integer(ik)	num_parameters
	The total number of parameters. More...

class(layercontainer), dimension(:), allocatable	list_layers
	The list of layers. More...

integer(ik), dimension(:), allocatable	ip_start
	The starting indices for the parameter repartition. More...

integer(ik), dimension(:), allocatable	ip_end
	The ending indices for the parameter repartition. More...

Private Attributes
integer(ik)	batch_size
	The batch size. More...

Detailed Description

Implements a sequential neural network.

Member Function/Subroutine Documentation

◆ apply_adjoint()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::apply_adjoint

Applies the adjoint of the networ. Implemented by snn_apply_adjoint.

◆ apply_forward()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::apply_forward

Applies and linearises the network. Implemented by snn_apply_forward.

◆ apply_tangent_linear()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::apply_tangent_linear

Applies the TL of the network. Implemented by snn_apply_tangent_linear.

◆ get_input_size()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::get_input_size

Returns the input size of the network. Implemented by snn_get_input_size.

◆ get_num_parameters()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::get_num_parameters

Returns the number of parameters. Implemented by snn_get_num_parameters.

◆ get_output_size()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::get_output_size

Returns the output size of the network. Implemented by snn_get_output_size.

◆ get_parameters()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::get_parameters

Getter for the network's parameters. Implemented by snn_get_parameters.

◆ set_parameters()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::set_parameters

Setter for the network's parameters. Implemented by snn_set_parameters.

◆ tofile()

procedure, pass, public fnn_network_sequential::sequentialneuralnetwork::tofile

Saves the network. Implemented by snn_tofile.

Member Data Documentation

◆ batch_size

integer(ik) fnn_network_sequential::sequentialneuralnetwork::batch_size

private

The batch size.

◆ ip_end

integer(ik), dimension(:), allocatable fnn_network_sequential::sequentialneuralnetwork::ip_end

The ending indices for the parameter repartition.

See sequentialneuralnetwork::ip_start.

◆ ip_start

integer(ik), dimension(:), allocatable fnn_network_sequential::sequentialneuralnetwork::ip_start

The starting indices for the parameter repartition.

If $\mathbf{p}$ is an array of size sequentialneuralnetwork::num_parameters containing the parameters for all layers, then the slice of $\mathbf{p}$ starting at ip_start(i) and ending at ip_end(i) contains the parameters for the i-th layer.

◆ list_layers

class(layercontainer), dimension(:), allocatable fnn_network_sequential::sequentialneuralnetwork::list_layers

The list of layers.

◆ num_layers

integer(ik) fnn_network_sequential::sequentialneuralnetwork::num_layers

The number of layers.

◆ num_parameters

integer(ik) fnn_network_sequential::sequentialneuralnetwork::num_parameters

The total number of parameters.

The documentation for this type was generated from the following file:

src/fnn_network_sequential.f90

Public Member Functions

Public Attributes

Private Attributes