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, 21 insertions, 0 deletions

diff --git a/sourcecodes/bnt-master/KPMtools/max_mult.m b/sourcecodes/bnt-master/KPMtools/max_mult.m
new file mode 100644
index 00000000..fab37e3a
--- /dev/null
+++ b/sourcecodes/bnt-master/KPMtools/max_mult.m
@@ -0,0 +1,21 @@
+function y=max_mult(A,x)
+% MAX_MULT Like matrix multiplication, but sum gets replaced by max
+% function y=max_mult(A,x) y(i) = max_j A(i,j) x(j)
+
+%X=ones(size(A,1),1) * x(:)'; % X(j,i) = x(i)
+%y=max(A.*X, [], 2);
+
+% This is faster
+if size(x,2)==1
+  X=x*ones(1,size(A,1)); % X(i,j) = x(i)
+  y=max(A'.*X)';
+else
+%this works for arbitrarily sized A and x (but is ugly, and slower than above)
+  X=repmat(x, [1 1 size(A,1)]);
+  B=repmat(A, [1 1 size(x,2)]);
+  C=permute(B,[2 3 1]);
+  y=permute(max(C.*X),[3 2 1]);
+%  this is even slower, as is using squeeze instead of permute
+%  Y=permute(X, [3 1 2]);
+%  y=permute(max(Y.*B, [], 2), [1 3 2]);
+end