Daniel@0: Daniel@0:
Daniel@0:Daniel@0: net = gtm(dimlatent, nlatent, dimdata, ncentres, rbfunc) Daniel@0: net = gtm(dimlatent, nlatent, dimdata, ncentres, rbfunc, prior) Daniel@0:Daniel@0: Daniel@0: Daniel@0:
net = gtm(dimlatent, nlatent, dimdata, ncentres, rbfunc)
,
Daniel@0: takes the dimension of the latent space dimlatent
, the
Daniel@0: number of data points sampled in the latent space nlatent
, the
Daniel@0: dimension of the data space dimdata
, the number of centres in the
Daniel@0: RBF model ncentres
, the activation function for the RBF
Daniel@0: rbfunc
Daniel@0: and returns a data structure net
. The parameters in the
Daniel@0: RBF and GMM sub-models are set by calls to the corresponding creation routines
Daniel@0: rbf
and gmm
.
Daniel@0:
Daniel@0:
The fields in net
are
Daniel@0:
Daniel@0: type = 'gtm' Daniel@0: nin = dimension of data space Daniel@0: dimlatent = dimension of latent space Daniel@0: rbfnet = RBF network data structure Daniel@0: gmmnet = GMM data structure Daniel@0: X = sample of latent points Daniel@0:Daniel@0: Daniel@0: Daniel@0:
net = gtm(dimlatent, nlatent, dimdata, ncentres, rbfunc, prior)
,
Daniel@0: sets a Gaussian zero mean prior on the
Daniel@0: parameters of the RBF model. prior
must be a scalar and represents
Daniel@0: the inverse variance of the prior distribution. This gives rise to
Daniel@0: a weight decay term in the error function.
Daniel@0:
Daniel@0:
gtmfwd
, gtmpost
, rbf
, gmm
Copyright (c) Ian T Nabney (1996-9) Daniel@0: Daniel@0: Daniel@0: Daniel@0: