BNW using Octave instead of Matlab.

This version of BNW should perform the same as the original version. The only difference is that it uses Octave instead of Matlab when running BayesNet Toolbox during parameter learning.

I am calling this BNW_1.02. It can be accessed at:
compbio.uthsc.edu/BNW_1.02

1 files changed, 27 insertions, 0 deletions

diff --git a/sourcecodes/bnt-master/KPMstats/condgaussTrainObserved.m b/sourcecodes/bnt-master/KPMstats/condgaussTrainObserved.m
new file mode 100644
index 00000000..2dd4cafa
--- /dev/null
+++ b/sourcecodes/bnt-master/KPMstats/condgaussTrainObserved.m
@@ -0,0 +1,27 @@
+function [mu, Sigma] = mixgaussTrainObserved(obsData, hiddenData, nstates, varargin);
+% mixgaussTrainObserved Max likelihood estimates of conditional Gaussian from raw data
+% function [mu, Sigma] = mixgaussTrainObserved(obsData, hiddenData, nstates, ...);
+%
+% Input:
+% obsData(:,i)
+% hiddenData(i)  - this is the mixture component label for example i
+% Optional arguments - same as mixgauss_Mstep
+%
+% Output:
+% mu(:,q)
+% Sigma(:,:,q) - same as mixgauss_Mstep
+
+[D numex] = size(obsData);
+Y = zeros(D, nstates);
+YY = zeros(D,D,nstates);
+YTY = zeros(nstates,1);
+w = zeros(nstates, 1);
+for q=1:nstates
+  ndx = find(hiddenData==q);
+  w(q) = length(ndx); % each data point has probability 1 of being in this cluster
+  data = obsData(:,ndx);
+  Y(:,q) = sum(data,2);
+  YY(:,:,q) = data*data';
+  YTY(q) = sum(diag(data'*data));
+end
+[mu, Sigma] = mixgauss_Mstep(w, Y, YY, YTY, varargin{:});