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, 22 insertions, 0 deletions

diff --git a/sourcecodes/bnt-master/SLP/misc/pdag_unsigned_to_signed.m b/sourcecodes/bnt-master/SLP/misc/pdag_unsigned_to_signed.m
new file mode 100644
index 00000000..37b414f1
--- /dev/null
+++ b/sourcecodes/bnt-master/SLP/misc/pdag_unsigned_to_signed.m
@@ -0,0 +1,22 @@
+function signed_pdag = pdag_unsigned_to_signed(pdag)
+
+% Input: pdag with '1' in one place for every directed edge and 
+%                  '1' in two places for every undirected edge
+% 
+% Output: pdag with '-1' in one place for every directed edge
+%                   '1' in two places for every undirected edge
+%
+% This function is used by Markov_equivalent_dags(dag) to convert 
+% output of SLP function dag_to_cpdag(dag) 
+% to format required as input for pdag_to_all_dags(pdag).
+%
+% I'm sure there's a prettier way to code this!
+%
+% Imme Ebert-Uphoff (ebert@tree.com), 2007
+%
+
+  undirected = ( (pdag+pdag')/2 == 1); % extract undirected eges
+  directed   = pdag - undirected;       % extract directed edges
+  signed_pdag = undirected - directed; % 1s for undirected, (-1)s for directed
+
+