edge_optimisation_function: More detailed edge fitting than mere cost_function.

In admixturegraph: Admixture Graph Manipulation and Fitting

Description

Returning the cost, an example edge solution of an optimal fit, and linear relations describing the set of all edge solutions. Operating with the full edge optimisation matrix, not the column reduced one.

Usage

edge_optimisation_function(data, concentration, matrix, graph,
  parameters = extract_graph_parameters(graph), iteration_multiplier = 3)

Arguments

data	The data set.
concentration	The Cholesky decomposition of the inverted covariance matrix.
matrix	A full edge optimisation matrix (typically given by the function build_edge_optimisation_matrix).
graph	The admixture graph.
parameters	In case one wants to tweak something in the graph.
iteration_multiplier	Given to mynonneg.

Value

Given an input vector of admix variables, returns a list containing the minimal error (cost), the graph-f4 statistics (approximation), an example solution (edge_fit), linear relations describing all the solutions (homogeneous) and one way to choose the free (free_edges) and bounded (bounded_edges) edge variables.

See Also

mynonneg

cost_function

log_likelihood

admixturegraph documentation built on May 2, 2019, 6:02 a.m.