Search found 1417 matches

by shooltz[BayesFusion]
Tue Mar 26, 2024 1:40 pm
Forum: GeNIe
Topic: Genie 2 vs 4 - discretization triggered
Replies: 4
Views: 11

Re: Genie 2 vs 4 - discretization triggered

Answered using private message.
by shooltz[BayesFusion]
Tue Mar 26, 2024 1:23 pm
Forum: GeNIe
Topic: Genie 2 vs 4 - discretization triggered
Replies: 4
Views: 11

Re: Genie 2 vs 4 - discretization triggered

Send me a private message on the forum, or an email to BayesFusion’s support account, Just to clarify the conditions which trigger discretization: an evidence set on any node with parents requires the discretization of the entire network. the evidence is set on a parent node only, not on any child n...
by shooltz[BayesFusion]
Tue Mar 26, 2024 12:55 pm
Forum: GeNIe
Topic: Genie 2 vs 4 - discretization triggered
Replies: 4
Views: 11

Re: Genie 2 vs 4 - discretization triggered

In the continuous or hybrid model, the discretization should be triggered when there's evidence in the nodes with parents. Can you share your network and the evidence which does not cause discretization in GeNIe 2.4? Without this information we cannot determine if this is a bug in old GeNIe 2.4, or ...
by shooltz[BayesFusion]
Tue Mar 26, 2024 12:53 pm
Forum: SMILE
Topic: Multiple genie versions on my PC
Replies: 2
Views: 6

Re: Multiple genie versions on my PC

You can have multiple GeNIe versions installed on the same computer.

The SMILE (and its wrappers, including rSMILE) do not depend on GeNIe, and their multiple versions can be installed on the same machine.
by shooltz[BayesFusion]
Mon Mar 04, 2024 7:32 pm
Forum: SMILE
Topic: Calculating values in Python and returning them to pySMILE
Replies: 10
Views: 2374

Re: Calculating values in Python and returning them to pySMILE

Unfortunately, currently we don't have an option to defined an 'external function' in SMILE. By external function I mean SMILE calling into user-specified code, which processes inputs (in practice, a current sample value from parent node(s)), and returns a value which is treated by SMILE in the same...
by shooltz[BayesFusion]
Thu Feb 22, 2024 9:37 pm
Forum: SMILE
Topic: Calculating values in Python and returning them to pySMILE
Replies: 10
Views: 2374

Re: Calculating values in Python and returning them to pySMILE

Just to make sure we're on the same wavelength - what of the two options below describes your goal: a) you want to use stress_list in the same 'system.xdsl' model loaded into your Python program running PySMILE or b) you want to derive some SMILE-compatible expression to use in an equation node in o...
by shooltz[BayesFusion]
Tue Feb 20, 2024 8:25 pm
Forum: SMILE
Topic: Calculating values in Python and returning them to pySMILE
Replies: 10
Views: 2374

Re: Calculating values in Python and returning them to pySMILE

What values are is your stress_list? Depending on its structure, the form of the SMILE equation would be different. For example, if the list can be approximated as histogram, you could use the Steps function. Can you post actual values from stress_list? If it's a very long list, please limit it to f...
by shooltz[BayesFusion]
Wed Feb 07, 2024 4:28 pm
Forum: SMILE
Topic: Calculating values in Python and returning them to pySMILE
Replies: 10
Views: 2374

Re: Calculating values in Python and returning them to pySMILE

Is stress_list conditional on any of the nodes in your SMILE model?
by shooltz[BayesFusion]
Wed Jan 24, 2024 7:37 pm
Forum: SMILE
Topic: is there a function for calculating the log likelihood?
Replies: 1
Views: 1774

Re: is there a function for calculating the log likelihood?

Sorry for the late reply. The most recent GeNIe build, version 4.1.3724, has a command which calculates the log likelihood for the network/dataset combination. This is implemented by calculating the sum of log(P(E)) values for each record in the dataset. SMILE will support this functionality in its ...
by shooltz[BayesFusion]
Wed Jan 24, 2024 7:35 pm
Forum: GeNIe
Topic: Genie shuts down when I attempt to run a sensitivity analysis
Replies: 2
Views: 883

Re: Genie shuts down when I attempt to run a sensitivity analysis

We have just released GeNIe 4.1.3724, which fixes the crash.

Please note that influence diagrams without a single terminal node have their utilities and sensitivity calculated as if there was a single child ALU node connected to all terminal utilities.
by shooltz[BayesFusion]
Mon Jan 22, 2024 3:26 pm
Forum: GeNIe
Topic: Genie shuts down when I attempt to run a sensitivity analysis
Replies: 2
Views: 883

Re: Genie shuts down when I attempt to run a sensitivity analysis

Crash confirmed, we're looking into it now.
by shooltz[BayesFusion]
Mon Jan 22, 2024 3:25 pm
Forum: SMILE
Topic: error with log likelihood during EM
Replies: 10
Views: 2138

Re: error with log likelihood during EM

The messages you're getting show that your data is inconsistent with the parameters in the network. You have a bunch of deterministic nodes, and you're trying to input an impossible combination of evidence from the data file.
by shooltz[BayesFusion]
Fri Jan 12, 2024 9:51 pm
Forum: SMILE
Topic: error with log likelihood during EM
Replies: 10
Views: 2138

Re: error with log likelihood during EM

In the em file I generated CPTs randomly If you don't have any initial probability distribution for your model, just use EM's internal randomization (call em.SetRandomizeParameters(true), and optionally fix the random seed for the initial parameters with em.SetSeed before SetRandomizeParameters). D...
by shooltz[BayesFusion]
Fri Jan 12, 2024 11:56 am
Forum: SMILE
Topic: error with log likelihood during EM
Replies: 10
Views: 2138

Re: error with log likelihood during EM

Your code will run fine when starting with ex7_TD2_10.xdsl. The log likelihood is around -990. If the model used is em_ex7_TD2_10_ED2_0.xdsl, the DSL_em::Learn will return DSL_ZERO_POTENTIAL, which is defined as -43. This is due to conflicting evidence in the dataset. Your code does not check for th...
by shooltz[BayesFusion]
Thu Jan 11, 2024 3:42 pm
Forum: SMILE
Topic: error with log likelihood during EM
Replies: 10
Views: 2138

Re: error with log likelihood during EM

We checked the log likelihood obtained after learning parameters with your files, and got values close to -1000 (in GeNIe and with a simple Python program.)

Maybe you did not create a proper matching between your network and data? Can you post your code fragment where you call EM?