Search found 41 matches

by jonnie
Sat Jun 02, 2012 5:39 pm
Forum: GeNIe
Topic: Diagnosis: Make Bipartite Graph crashes
Replies: 4
Views: 4650

Re: Diagnosis: Make Bipartite Graph crashes

Haha, I guess that network is simply not meant for bipartition! Well I've looked on my task manager and I didn't see any memory spike before GeNIe went down, so I figured that can't be the reason... Maybe it would be possible if GeNIe could generate a noisyMAX bipartite graph, but then again, this w...
by jonnie
Sat Jun 02, 2012 3:38 pm
Forum: GeNIe
Topic: Diagnosis: Make Bipartite Graph crashes
Replies: 4
Views: 4650

Re: Diagnosis: Make Bipartite Graph crashes

Yeah, works! Awesome bugfix time :) Now I'm sorry, but making a bipartite graph still crashes when I want to convert the whole network... The crash looks different this time. I'll send you the problematic network via PM. The crash is immediate so I guess GeNIe is not running out of computer resource...
by jonnie
Sat Jun 02, 2012 12:22 pm
Forum: GeNIe
Topic: Diagnosis: Make Bipartite Graph crashes
Replies: 4
Views: 4650

Diagnosis: Make Bipartite Graph crashes

Hello I found a new nut for you to crack :) I have this example network here and it's annotated with diagnostic information and I want to create a bipartite graph out of it! Unfortunately, GeNIe crashes. But it doesn't crash when I remove the yellow node (by deletion or marginalization doesn't matte...
by jonnie
Thu May 24, 2012 1:19 pm
Forum: SMILE
Topic: Quiet failures in setting evidence / updating beliefs
Replies: 6
Views: 7695

Re: Quiet failures in setting evidence / updating beliefs

Hello, I stumbled upon another phenomenon similar to this: LBP fails quietly. The network is no polytree so Pearl fails of course. However, I expected LBP to perform fine, since it's not really a sampling algorithm. The problem with sampling algorithms is that my network contains lots of determinist...
by jonnie
Thu May 24, 2012 1:00 pm
Forum: GeNIe
Topic: GeNIe feature request: undo
Replies: 1
Views: 3078

GeNIe feature request: undo

If you're thinking about adding new features to GeNIe (like the finding of unnecessary arcs recently), I'd like to suggest and undo/redo function :D I know it's not a trivial thing to implement; an undo/redo functionality requires classes for all the different actions a user can perform on the netwo...
by jonnie
Wed May 16, 2012 5:26 pm
Forum: GeNIe
Topic: Genie crashes on this network ... why?
Replies: 7
Views: 5825

Re: Genie crashes on this network ... why?

Quite courageous of you to form a circle in a BN! But seems it works :)
Some more things you could try (if you have the time to do so manually) is to define the nodes with noisy-MAX parameters, and to divorce them. That really works wonders sometimes :)
by jonnie
Mon May 14, 2012 4:28 pm
Forum: SMILE
Topic: Relevance Reasoning explained
Replies: 6
Views: 6912

Re: Relevance Reasoning explained

Good evening, I just wanted to say thanks to your detailed & exhaustive answers! These things really explain a lot of the behavior I'm observing and will help a lot in drawing conclusions as well. Greetings, Johannes Update: I measure the times of the different steps separately. I was already wo...
by jonnie
Fri May 11, 2012 8:56 am
Forum: SMILE
Topic: strengths field in noisyMax node
Replies: 3
Views: 4324

Re: strengths field in noisyMax node

Combining chained noisy-OR nodes?? Well, I'm not the programmer of SamIam, but that seems like a weird thing to do. Will it also "chain" them again when it saves them? I've made some research about the effect of parent divorcing on inference time (complexity, stability), and I have to say,...
by jonnie
Thu May 10, 2012 3:14 pm
Forum: SMILE
Topic: Relevance Reasoning explained
Replies: 6
Views: 6912

Re: Relevance Reasoning explained

Thanks a lot! I'm sorry but I came up with a whole bunch of follow-up questions :D 'relevance' has two meanings in SMILE: a) the code which propagates evidence and invalidates node values when network structure and/or evidence set changes, and b) the code which simplifies network before running actu...
by jonnie
Thu May 10, 2012 9:27 am
Forum: SMILE
Topic: Bug? Assertion fails when relevance reasoning disabled
Replies: 11
Views: 9576

Re: Bug? Assertion fails when relevance reasoning disabled

Btw I noticed that the assertion also fails when my relevance flags are set like this: SetRelevanceLevelFlags(DSL_RELEVANCE_EVIDENCE_SET || DSL_RELEVANCE_EVIDENCE_CLEARED || DSL_RELEVANCE_PROPAGATED_EVIDENCE_SET || DSL_RELEVANCE_PROPAGATED_EVIDENCE_CLEARED || DSL_RELEVANCE_GENERAL_CHANGE) Haha yeah...
by jonnie
Thu May 10, 2012 9:23 am
Forum: SMILE
Topic: Relevance Reasoning explained
Replies: 6
Views: 6912

Relevance Reasoning explained

Hello DSL, I'd like to know a little bit more about how and where you do relevance reasoning. I know the flags in network.h and I guess that for example relevance reasoning related to evidence will be executed during the call of SetEvidence. - If I set evidence and activate relevance reasoning after...
by jonnie
Mon May 07, 2012 9:02 am
Forum: SMILE
Topic: strengths field in noisyMax node
Replies: 3
Views: 4324

Re: strengths field in noisyMax node

The explanation text in the wiki certainly needs some polishing! The English is incredibly bad! :P But I think the diagram explains quite well how to use them. There's also an explanation for GeNIe that might be more understandable: http://genie.sis.pitt.edu/GeNIeHelp/Elements_of_GeNIe/Diagrams/Nois...
by jonnie
Wed May 02, 2012 12:48 pm
Forum: SMILE
Topic: Calculating a joint probability distribution
Replies: 1
Views: 5817

Calculating a joint probability distribution

I have a bayesian network where the nodes are designated to be either disease nodes, symptom nodes, or auxiliary. Given evidence on some of the symptoms, I want to calculate the *joint* probability distribution of the disease nodes. That means not diseases' marginals but the probabilities of their c...
by jonnie
Thu Apr 26, 2012 12:57 pm
Forum: SMILE
Topic: Marginalize node: GeNIe vs. Smile
Replies: 5
Views: 5806

Re: Marginalize node: GeNIe vs. Smile

Thanks a lot for clarifying. I have a few follow-up questions: - In order to preserve the JPD, the children of the marginalized node are "bound together" by adding arcs between them. Does this always result in a complete subgraph of the children? - Is there any way to know which children w...
by jonnie
Wed Apr 25, 2012 9:02 am
Forum: SMILE
Topic: Marginalize node: GeNIe vs. Smile
Replies: 5
Views: 5806

Re: Marginalize node: GeNIe vs. Smile

It should work regarless of the child count of the marginalized node. Seems I interpreted the wrong thing into GeNIe's behavior. I was thinking that it's meant to be like that, that you need a single child like in DSL_network::MarginalizeNode(int, int). I've reproduced marginalizing with several ch...