Search found 1179 matches

by shooltz[BayesFusion]
Mon Jun 24, 2019 1:56 pm
Forum: GeNIe
Topic: Most determinant nodes
Replies: 4
Views: 22

Re: Most determinant nodes

You can consider using sensitivity analysis. It will give you the measure of the influence of each parameter of the network on the posterior probability of the selected target node and its outcome.
by shooltz[BayesFusion]
Mon Jun 17, 2019 3:52 pm
Forum: GeNIe
Topic: GeNIe 2.4 released
Replies: 0
Views: 11

GeNIe 2.4 released

GeNIe 2.4 was just released. In this version we added, among other things, the following:

- user-defined functions
- Poisson() distribution
- Erf() function
- support for noisyMAX/noisyAdder nodes in hybrid models
- access to models stored in BayesBox
by shooltz[BayesFusion]
Fri Apr 19, 2019 6:14 pm
Forum: GeNIe
Topic: BayesBox and BayesMobile released
Replies: 0
Views: 304

BayesBox and BayesMobile released

We have released two new products: BayesBox : interactive network repository/dashboard accessed through any web browser. Online demo is available at https://demo.bayesfusion.com BayesMobile : diagnostic GeNIe/SMILE models on a mobile device. It's now available in the Apple AppStore; Android version ...
by shooltz[BayesFusion]
Tue Mar 19, 2019 2:06 pm
Forum: GeNIe
Topic: Submodel - a question + a suggestion
Replies: 4
Views: 284

Re: Submodel - a question + a suggestion

The result: those relocated nodes just stay there in coexistence with the submodel node that previously represents them.
Dragging nodes between submodels works for me. See the attached animated gif; are you referring to some other drag'n'drop action?
by shooltz[BayesFusion]
Thu Feb 28, 2019 8:51 pm
Forum: GeNIe
Topic: To make a truncated distribution
Replies: 5
Views: 432

Re: To make a truncated distribution

GeNIe 2.3.3828 (available now) fixes the issues reported in this forum thread.
by shooltz[BayesFusion]
Thu Feb 28, 2019 1:00 pm
Forum: SMILE
Topic: Help with Bayesian Modeling
Replies: 6
Views: 1028

Re: Help with Bayesian Modeling

The answer from the StackOverflow thread is correct; Smile.NET is a mixed mode DLL which contains both native Intel code and Microsoft Intermediate language. This type of DLL is not directly supported by Unity.
by shooltz[BayesFusion]
Fri Feb 22, 2019 9:57 pm
Forum: GeNIe
Topic: Any idea to create a truncated distribution?
Replies: 7
Views: 462

Re: Any idea to create a truncated distribution?

The TruncNormal distribution generates samples by drawing a normal sample specified by its two first arguments (mean, stddev), then compares it agains the third argument (the cutoff). If the value is greater than cutoff, the sample is accepted, otherwise the procedure is repeated. To avoid infinite ...
by shooltz[BayesFusion]
Fri Feb 22, 2019 11:43 am
Forum: GeNIe
Topic: A simple question
Replies: 2
Views: 226

Re: A simple question

The Elicitate command opens the interactive chart where you can move the interval boundaries with a mouse. The CPT is then adjusted accordingly. Note that GeNIe does not store the discretized CPTs in the xdsl files - only the intervals. The CPTs are derived at runtime. We'll fix the GeNIe manual to ...
by shooltz[BayesFusion]
Fri Feb 22, 2019 11:37 am
Forum: GeNIe
Topic: Discretization problem
Replies: 2
Views: 229

Re: Discretization problem

During the discretization process the CPTs for the equation nodes are derived by sampling. Some of these samples may fall outside of the bounds specified for the node. If the sample value is less than a lower node bound, it's rejected and a count of underflow samples is increased. If the sample valu...
by shooltz[BayesFusion]
Fri Feb 22, 2019 11:30 am
Forum: GeNIe
Topic: Any idea to create a truncated distribution?
Replies: 7
Views: 462

Re: Any idea to create a truncated distribution?

I want to create a truncated distribution - e.g. a normal distribution N(5,1) be truncated at a threshold, say 3, so that there is no frequency of occurrence before the threshold (3). Making it bit more complex is that the threshold itself is stochastic subjected to a distribution (rather than a fi...
by shooltz[BayesFusion]
Tue Feb 05, 2019 7:44 pm
Forum: GeNIe
Topic: User-defined PDF function
Replies: 2
Views: 292

Re: User-defined PDF function

The CustomPDF function requires the x coordinates (the first half of function arguments) to be in increasing order. Your equations have descreasing x ordering.

BTW, please check your private message inbox on this forum.
by shooltz[BayesFusion]
Tue Jan 29, 2019 7:11 pm
Forum: GeNIe
Topic: Hybrid model program
Replies: 2
Views: 261

Re: Hybrid model program

That is to say the network cannot contain any decision and value node. At this point we don't support the equation nodes in influence diagrams - it would require relatively large amount of work. Note that you can use expressions in MAU utility nodes, including expressions conditioned by decision pa...
by shooltz[BayesFusion]
Mon Jan 21, 2019 8:36 pm
Forum: GeNIe
Topic: Urgent - errors in LEAK column
Replies: 11
Views: 1316

Re: Urgent - errors in LEAK column

Please check your private messages.
by shooltz[BayesFusion]
Mon Jan 21, 2019 7:13 pm
Forum: SMILE
Topic: Problem with structure learning
Replies: 5
Views: 514

Re: Problem with structure learning

I tried my dataset with GeNIe and Greedy Thick Thinning algorithm and it works pretty well
Your data file contains columns with missing values, which are not compatible with Greedy Thick Thinning. Unless you've switched to another data file, GeNIe should tell you about the problems with your data.
by shooltz[BayesFusion]
Fri Jan 18, 2019 1:16 pm
Forum: SMILE
Topic: Problem with structure learning
Replies: 5
Views: 514

Re: Problem with structure learning

Naive Bayes algorithm needs to know which variable in your dataset is a class variable. In SMILE.NET you can specify the class variable by setting the NaiveBayes.ClassVariableId property. The class variable identifier should be matched by one of the variables in the dataset passed as a parameter to ...