EVOLUTION-MANAGER
Edit File: SOM.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><title>R: Self-Organizing Maps: Online Algorithm</title> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <link rel="stylesheet" type="text/css" href="R.css" /> </head><body> <table width="100%" summary="page for SOM {class}"><tr><td>SOM {class}</td><td style="text-align: right;">R Documentation</td></tr></table> <h2> Self-Organizing Maps: Online Algorithm </h2> <h3>Description</h3> <p>Kohonen's Self-Organizing Maps are a crude form of multidimensional scaling. </p> <h3>Usage</h3> <pre> SOM(data, grid = somgrid(), rlen = 10000, alpha, radii, init) </pre> <h3>Arguments</h3> <table summary="R argblock"> <tr valign="top"><td><code>data</code></td> <td> <p>a matrix or data frame of observations, scaled so that Euclidean distance is appropriate. </p> </td></tr> <tr valign="top"><td><code>grid</code></td> <td> <p>A grid for the representatives: see <code><a href="somgrid.html">somgrid</a></code>. </p> </td></tr> <tr valign="top"><td><code>rlen</code></td> <td> <p>the number of updates: used only in the defaults for <code>alpha</code> and <code>radii</code>. </p> </td></tr> <tr valign="top"><td><code>alpha</code></td> <td> <p>the amount of change: one update is done for each element of <code>alpha</code>. Default is to decline linearly from 0.05 to 0 over <code>rlen</code> updates. </p> </td></tr> <tr valign="top"><td><code>radii</code></td> <td> <p>the radii of the neighbourhood to be used for each update: must be the same length as <code>alpha</code>. Default is to decline linearly from 4 to 1 over <code>rlen</code> updates. </p> </td></tr> <tr valign="top"><td><code>init</code></td> <td> <p>the initial representatives. If missing, chosen (without replacement) randomly from <code>data</code>. </p> </td></tr></table> <h3>Details</h3> <p><code>alpha</code> and <code>radii</code> can also be lists, in which case each component is used in turn, allowing two- or more phase training. </p> <h3>Value</h3> <p>An object of class <code>"SOM"</code> with components </p> <table summary="R valueblock"> <tr valign="top"><td><code>grid</code></td> <td> <p>the grid, an object of class <code>"somgrid"</code>. </p> </td></tr> <tr valign="top"><td><code>codes</code></td> <td> <p>a matrix of representatives. </p> </td></tr></table> <h3>References</h3> <p>Kohonen, T. (1995) <em>Self-Organizing Maps.</em> Springer-Verlag </p> <p>Kohonen, T., Hynninen, J., Kangas, J. and Laaksonen, J. (1996) <em>SOM PAK: The self-organizing map program package.</em> Laboratory of Computer and Information Science, Helsinki University of Technology, Technical Report A31. </p> <p>Ripley, B. D. (1996) <em>Pattern Recognition and Neural Networks.</em> Cambridge. </p> <p>Venables, W. N. and Ripley, B. D. (2002) <em>Modern Applied Statistics with S.</em> Fourth edition. Springer. </p> <h3>See Also</h3> <p><code><a href="somgrid.html">somgrid</a></code>, <code><a href="batchSOM.html">batchSOM</a></code> </p> <h3>Examples</h3> <pre> require(graphics) data(crabs, package = "MASS") lcrabs <- log(crabs[, 4:8]) crabs.grp <- factor(c("B", "b", "O", "o")[rep(1:4, rep(50,4))]) gr <- somgrid(topo = "hexagonal") crabs.som <- SOM(lcrabs, gr) plot(crabs.som) ## 2-phase training crabs.som2 <- SOM(lcrabs, gr, alpha = list(seq(0.05, 0, len = 1e4), seq(0.02, 0, len = 1e5)), radii = list(seq(8, 1, len = 1e4), seq(4, 1, len = 1e5))) plot(crabs.som2) </pre> <hr /><div style="text-align: center;">[Package <em>class</em> version 7.3-15 <a href="00Index.html">Index</a>]</div> </body></html>