BayesFusion Support Forum

ckerse wrote:Can I run jsmile on a mac? I've been a windows user and am having trouble porting the code to the mac.
Thanks!

We don't provide jSMILE binaries for OSX, but the source code is available. You just have to compile it.

However I still get the same error. I also tried commenting out the "DSL_dataset ds;" after which the code then compiled correctly. I don't know if there was a form of instillation (./configure?) that I maybe missed.

There's no install/setup - unpacking the files is enough.

What happens if you ...

You have a space in $(SMILE) - it's between 'Smile\' and 'trial'.

Can you post your gcc command line here?

You'll need to tell the linker to use smile.lib (automatic linking is only supported in Visual C++). I've never used CodeBlocks, so I can't give you more assistance, other than pointing to section 1.11.8 in the CodeBlocks online documentation.

Are you sure that you have most recent jSMILE version? Currently, the Network class does not contain the 'newNetwork' method (which caused the UnsatisfiedLinkError exception to be thrown).

ystf wrote:The problem is that my program is kind of huge, with many modules and so on and also some private data...

In such case, try to reproduce the crash with smaller application.

I won't be able to help without your Java source code.

I read the post on the Indexing parents, but I fail to understand how it relates to my question. Indexing parents are parents , not the predecessors, so I don't see how to make the connection.

The indexing parents set may contain predecessors - check the definition. If you have GeNIe installed ...

nikkne wrote:Example iterates through the list of indexing parents, which is enough if the network/ID is of height 1. What to do when the network is much, much bigger (for example, there are 5 chance nodes between a decision and value node)?

Check my reply to your post about indexing parents.

Here's the definition of 'indexing parents' taken from http://genie.sis.pitt.edu/wiki/Brief_Introduction_to_Some_Simple_Classes:_Solving_Influence_Diagrams:

Indexing Parents
The results for each node will be indexed by its indexing parents. These indexing parents are the decision nodes and ...

Is it possible to introduce in the Genie sw this kind of Monte-Carlo simulation, obtaining a spreadsheet with the results of the different runs?

We don't plan to implement such feature at the moment. We usually encourage users to write some code using core SMILE or one of the wrappers to perform ...

Am I missing something?

You can use DSL_ALG_ID_SHACHTER algorithm - it'll find the best policy at SMILE level. It's the equivalent of 'Find best policy' selection from the Network|Algorithm in GeNIe. The code in the tutorial invokes the default influence diagram algorithm DSL_ALG_ID_COOPERSOLVING ...

I've trained my network with a trainings set.
Now I want to test it with the test set.
Is there a possibility to import the complete test set and compute it or have I to choose the evidence for each node by hand?

The closest thing would be using Data|Copy Cases command, which would create the ...

RemoveOutcome will fail if you have just two states/choices - SMILE always keeps the network in consistent state, and discrete node with just one outcome certainly breaks the consistency.

SetNumberOfOutcomes will fail if node has children. This is rather obscure method, mainly used by file I/O ...