Mercurial > pylearn

comparison pylearn/algorithms/mcRBM.py @ 985:78b5bdf967f6

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
mcRBM - removed references to dropped 'P' matrix
author James Bergstra <bergstrj@iro.umontreal.ca>
date Tue, 24 Aug 2010 13:51:26 -0400
parents 5badf36a6daf
children bee0ca674b2b
comparison
equal deleted inserted replaced
984:5badf36a6daf 985:78b5bdf967f6
105 F squared filters, 105 F squared filters,
106 J mean variables, and 106 J mean variables, and
107 K covariance variables. 107 K covariance variables.
108 The mcRBM is parametrized by 5 variables: 108 The mcRBM is parametrized by 5 variables:
109 109
110 - `P`, a matrix (probably sparse) of pooling (F x K)
111 - `U`, a matrix whose rows are visible covariance directions (I x F) 110 - `U`, a matrix whose rows are visible covariance directions (I x F)
112 - `W`, a matrix whose rows are visible mean directions (I x J) 111 - `W`, a matrix whose rows are visible mean directions (I x J)
113 - `b`, a vector of hidden covariance biases (K) 112 - `b`, a vector of hidden covariance biases (K)
114 - `c`, a vector of hidden mean biases (J) 113 - `c`, a vector of hidden mean biases (J)
115 114
195 import logging 194 import logging
196 import numpy as np 195 import numpy as np
197 import numpy 196 import numpy
198 from theano import function, shared, dot 197 from theano import function, shared, dot
199 from theano import tensor as TT 198 from theano import tensor as TT
200 import theano.sparse #installs the sparse shared var handler
201 floatX = theano.config.floatX 199 floatX = theano.config.floatX
202 200
203 from pylearn.sampling.hmc import HMC_sampler 201 from pylearn.sampling.hmc import HMC_sampler
204 from pylearn.io import image_tiling 202 from pylearn.io import image_tiling
205 203