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, 86 insertions, 0 deletions

diff --git a/sourcecodes/bnt-master/BNT/inference/static/@belprop_inf_engine/enter_evidence.m b/sourcecodes/bnt-master/BNT/inference/static/@belprop_inf_engine/enter_evidence.m
new file mode 100644
index 00000000..88cce18e
--- /dev/null
+++ b/sourcecodes/bnt-master/BNT/inference/static/@belprop_inf_engine/enter_evidence.m
@@ -0,0 +1,86 @@
+function [engine, ll, niter] = enter_evidence(engine, evidence, varargin)
+% ENTER_EVIDENCE Propagate evidence using belief propagation
+% [engine, ll, niter] = enter_evidence(engine, evidence, ...)
+%
+% The log-likelihood is not computed; ll = 0.
+% niter contains the number of iterations used (if engine.protocol = 'parallel')
+%
+% The following optional arguments can be specified in the form of name/value pairs:
+% [default value in brackets]
+%
+% maximize - 1 means use max-product, 0 means use sum-product [0]
+% exclude  - list of nodes whose potential will not be included in the joint [ [] ]
+%
+% e.g., engine = enter_evidence(engine, ev, 'maximize', 1)
+
+ll = 0;
+exclude = [];
+maximize = 0;
+
+if nargin >= 3
+  args = varargin;
+  nargs = length(args);
+  for i=1:2:nargs
+    switch args{i},
+     case 'exclude', exclude = args{i+1};
+     case 'maximize', maximize = args{i+1};
+     otherwise,
+      error(['invalid argument name ' args{i}]);
+    end
+  end
+end
+
+engine.maximize = maximize;
+
+if ~isempty(engine.filename)
+  engine.fid = fopen(engine.filename, 'w');
+  if engine.fid == 0
+    error(['can''t open ' engine.filename]);
+  end
+else
+  engine.fid = [];
+end
+
+gdl = engine.gdl;
+bnet = bnet_from_engine(engine);
+
+ndoms = length(gdl.doms);
+ns = bnet.node_sizes;
+onodes = find(~isemptycell(evidence));
+pot_type = determine_pot_type(bnet, onodes);
+
+% prime each local kernel with evidence (if any)
+local_kernel = cell(1, ndoms);
+for i=1:ndoms
+  if myismember(i, exclude)
+    local_kernel{i} =  mk_initial_pot(pot_type, gdl.doms{i}, ns, bnet.cnodes, onodes);
+  else
+    e = bnet.equiv_class(i);
+    local_kernel{i} =  convert_to_pot(bnet.CPD{e}, pot_type, gdl.doms{i}(:), evidence);
+  end
+end
+  
+% initialise all msgs to 1s
+msg = cell(ndoms, ndoms);
+for i=1:ndoms
+  nbrs = gdl.nbrs{i};
+  for j=nbrs(:)'
+    dom = gdl.sepset{i,j};
+    msg{i,j} = mk_initial_pot(pot_type, dom, ns, bnet.cnodes, onodes);
+  end
+end
+
+switch engine.protocol
+ case 'parallel', 
+   [engine.marginal_domains, niter] = parallel_protocol(engine, evidence, pot_type, local_kernel, msg);
+ case 'tree',
+  engine.marginal_domains = serial_protocol(engine, evidence, pot_type, local_kernel, msg);
+  niter = 1;
+end
+engine.niter = niter;
+
+%fprintf('just finished %d iterations of belprop\n', niter);
+
+if ~isempty(engine.filename)
+  fclose(engine.fid);
+end