wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual gpfwd wolffd@0: wolffd@0: wolffd@0: wolffd@0:

gpfwd wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Forward propagation through Gaussian Process. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0:

wolffd@0: y = gpfwd(net, x)
wolffd@0: [y, sigsq] = gpfwd(net, x)
wolffd@0: [y, sigsq] = gpfwd(net, x, cninv)
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: y = gpfwd(net, x) takes a Gaussian Process data structure net wolffd@0: together wolffd@0: with a matrix x of input vectors, and forward propagates the inputs wolffd@0: through the model to generate a matrix y of output wolffd@0: vectors. Each row of x corresponds to one input vector and each wolffd@0: row of y corresponds to one output vector. This assumes that the wolffd@0: training data (both inputs and targets) has been stored in net by wolffd@0: a call to gpinit; these are needed to compute the training wolffd@0: data covariance matrix. wolffd@0: wolffd@0:

[y, sigsq] = gpfwd(net, x) also generates a column vector sigsq of wolffd@0: conditional variances (or squared error bars) where each value corresponds to a pattern. wolffd@0: wolffd@0:

[y, sigsq] = gpfwd(net, x, cninv) uses the pre-computed inverse covariance wolffd@0: matrix cninv in the forward propagation. This increases efficiency if wolffd@0: several calls to gpfwd are made. wolffd@0: wolffd@0:

wolffd@0: Example wolffd@0:

wolffd@0: The following code creates a Gaussian Process, trains it, and then plots the wolffd@0: predictions on a test set with one standard deviation error bars: wolffd@0:

wolffd@0: 
wolffd@0: net = gp(1, 'sqexp');
wolffd@0: net = gpinit(net, x, t);
wolffd@0: net = netopt(net, options, x, t, 'scg');
wolffd@0: [pred, sigsq] = gpfwd(net, xtest);
wolffd@0: plot(xtest, pred, '-k');
wolffd@0: hold on
wolffd@0: plot(xtest, pred+sqrt(sigsq), '-b', xtest, pred-sqrt(sigsq), '-b');
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: gp, demgp, gpinit

wolffd@0: Pages: wolffd@0: Index wolffd@0:

wolffd@0: