---

gnn_convergence : Input Convergence.
[Atomic Nodes.]

---

Detailed Description

The gnn_divergence : Input Divergence. node is a special node that takes a big input vector and slices it into mini-vectors that fit into the output. This can be viewed a if the input where in reality a concatenation of subvectors of the same size.

The function implemented by this node is given by:

In other words, it slices the input vector into several subvectors of size and sums them up, divided by the total amount of resulting pieces . Remember that is the output size.

If the input vector's size isn't a multiple of the output vector's size, that is, if can't be written as , then the last vector slice will consider only the remaining terms, e.g. if and is of size then

Schematically, this idea can be depicted as

The function's gradient is:

Functions
int	gnn_convergence_f (gnn_node *node, const gsl_vector *x, const gsl_vector *w, gsl_vector *y)
	Computes the output.
int	gnn_convergence_dx (gnn_node *node, const gsl_vector *x, const gsl_vector *w, const gsl_vector *dy, gsl_vector *dx)
	Computes .
int	gnn_convergence_dw (gnn_node *node, const gsl_vector *x, const gsl_vector *w, const gsl_vector *dy, gsl_vector *dw)
	Computes .
gnn_node *	gnn_convergence_new (size_t input_size, size_t output_size)
	Creates an input convergence node.

---

Function Documentation

int gnn_convergence_dw (  gnn_node *    node, const gsl_vector *    x, const gsl_vector *    w, const gsl_vector *    dy, gsl_vector *    dw )  [static]

Parameters: node  A pointer to the gnn_divergence : Input Divergence. node. x  The input vector . w  The parameter vector . dy  The vector . dx  The output buffer vector . Returns: Returns 0 on success. Definition at line 203 of file gnn_convergence.c.

int gnn_convergence_dx (  gnn_node *    node, const gsl_vector *    x, const gsl_vector *    w, const gsl_vector *    dy, gsl_vector *    dx )  [static]

Parameters: node  A pointer to the gnn_divergence : Input Divergence. node. x  The input vector . w  The parameter vector . dy  The vector . dx  The output buffer vector . Returns: Returns 0 on success. Definition at line 159 of file gnn_convergence.c.

int gnn_convergence_f (  gnn_node *    node, const gsl_vector *    x, const gsl_vector *    w, gsl_vector *    y )  [static]

Parameters: node  A pointer to the gnn_divergence : Input Divergence. node. x  The input vector . w  The parameter vector . y  The output buffer vector . Returns: Returns 0 on success. Definition at line 114 of file gnn_convergence.c.

gnn_node* gnn_convergence_new (  size_t    input_size, size_t    output_size )

This function creates a node of the gnn_divergence : Input Divergence. type. Parameters: input_size  The input size . output_size  The output size . Returns: A pointer to a new gnn_divergence : Input Divergence. node. Definition at line 229 of file gnn_convergence.c.

---