agraph_one_lambda

Segmentation using graph structure

Perform segmentation of graph-based signals. Assume a noisy observation
of a signal two values correspond to vertices on a graph. Assume the true
value of the signal is piece-wise constant (where each 'piece' is a
connected subgraph). The main function, agraph(), computes the
segmentation of the signal. The package also includes a wrapper around the competing
method flsa() (from package 'flsa').
More information about this method in Goepp and van de Kassteele (2022)
"Graph-Based Spatial Segmentation of
Health-Related Areal Data" <doi:10.48550/arXiv.2206.06752>.

Vivien Goepp

graphseg

Segmentation of Graph-Based Signals

agraph_one_lambda function

<dl><dt>gamma</dt>
<dd>entry vector to regularize</dd>
<dt>graph</dt>
<dd>an igraph object (from package <code>igraph</code>) giving the regularization structure</dd>
<dt>lambda</dt>
<dd>regularizing constant</dd>
<dt>weights</dt>
<dd>weights for gamma. Default value is one.</dd>
<dt>delta</dt>
<dd>Computational constant in the adaptive ridge reweighting formula.</dd>
<dt>tol</dt>
<dd>Tolerance to test for convergence of the adaptive ridge</dd>
<dt>thresh</dt>
<dd>Thresholding constant used to fuse two adjacent regions with close value of <code>gamma</code>.</dd></dl>

Arguments

Segmentation using graph structure — agraph_one_lambda

<dl>

<dt>gamma</dt>
<dd>entry vector to regularize</dd>


<dt>graph</dt>
<dd>an igraph object (from package <code>igraph</code>) giving the regularization structure</dd>


<dt>lambda</dt>
<dd>regularizing constant</dd>


<dt>weights</dt>
<dd>weights for gamma. Default value is one.</dd>


<dt>delta</dt>
<dd>Computational constant in the adaptive ridge reweighting formula.</dd>


<dt>tol</dt>
<dd>Tolerance to test for convergence of the adaptive ridge</dd>


<dt>thresh</dt>
<dd>Thresholding constant used to fuse two adjacent regions with close value of <code>gamma</code>.</dd>

</dl>

Get 50% off unlimited learning

agraph_one_lambda: Segmentation using graph structure

Description

Usage

Arguments