Search found 430 matches
Re: empty ROC
Dear Daniel, One error that I see in your calculation is that you assumed/calculated P(HD=yes)=0.39. Actually, P(HD=yes)=0.61 and P(HD=no)=0.39, so you may have made an error here. Does this solve your problem? I love to hear that you are going to use GeNIe in your research. GeNIe is incredibly popu...
- Wed Aug 16, 2017 8:02 pm
- Forum: GeNIe
- Topic: Advice please on canonical model
- Replies: 12
- Views: 8924
Re: Advice please on canonical model
Hi Charlie, The leak is present in CPTs as well. It is the probability of the effect happening when none of the parents are present. If you click the CPT button in a Noisy node, you will certainly see the leak! Interesting that you are having problems with the leak. I find it one of the most straigh...
- Tue Aug 15, 2017 6:42 am
- Forum: GeNIe
- Topic: Advice please on canonical model
- Replies: 12
- Views: 8924
Re: Advice please on canonical model
Hi Charlie, I'm sorry to have missed this (that you have already looked at the Diez & Druzdzel paper). My experience in applying canonical models is that they each require careful thinking. There is no easy/algorithmic solution to which model to pick. Am I understanding correctly that you have s...
- Mon Aug 14, 2017 11:47 am
- Forum: GeNIe
- Topic: Advice please on canonical model
- Replies: 12
- Views: 8924
Re: Advice please on canonical model
My suggestion is that you look carefully at the paper that I have co-authored with my colleague Javier Diez, available at the following location: http://www.cisiad.uned.es/techreports/canonical.php We have tried to review all existing canonical models. Perhaps you will find in there what you are loo...
Re: empty ROC
Dear Daniel, Having missing values in the class variable is generally odd -- you cannot test the record for accuracy, as you don't know what class the record belongs to. Quality software, like GeNIe should, however, behave correctly even in this case. We have fixed this and you can download the newe...
Re: empty ROC
Hi Daniel, I have looked at your model and your data and there is indeed something strange going on that has to do with records with missing class variable. Once you delete these from Validation, you will get both ROC and calibration curves. It seems GeNIe should be able to handle this. I will let y...
- Thu Jul 20, 2017 8:47 am
- Forum: GeNIe
- Topic: Validation process: model's result should be probability and not exact value (?)
- Replies: 3
- Views: 3709
Re: Validation process: model's result should be probability and not exact value (?)
I have just posted a reply to your questions in the original thread -- Marek
- Thu Jul 20, 2017 8:46 am
- Forum: GeNIe
- Topic: Ask learning process, fixed nodes, and validation process
- Replies: 9
- Views: 7012
Re: Ask learning process, fixed nodes, and validation process
1. When perform parameter learning, we get log(p) value. Do I need to consider this value to evaluate my model’s performance? I see that closer to zero is better. My log (p) value is -1084380.522721 which is very high. The absolute value of log(p) depends on the size of your network, data file and ...
- Fri Jul 07, 2017 2:55 pm
- Forum: GeNIe
- Topic: Ask learning process, fixed nodes, and validation process
- Replies: 9
- Views: 7012
Re: Ask learning process, fixed nodes, and validation process
1. Have you tried looking at the output file? You can see the posterior probability distributions over the class variable for each of your data records Sorry, I did not get this. What do you mean by looking at the output file? When you run validation, you can create an output file. It will replicat...
- Wed Jul 05, 2017 5:41 am
- Forum: GeNIe
- Topic: Ask learning process, fixed nodes, and validation process
- Replies: 9
- Views: 7012
Re: Ask learning process, fixed nodes, and validation process
Hi Daniel, But what is the function of "fixed nodes" in validation? You fix them if you do not want their CPTs to change during the cross-validation. They will normally change, because cross-validation includes a learning and a testing phase. 1. in the outcome node, I have only 2 states: &...
- Mon Jun 26, 2017 6:37 am
- Forum: GeNIe
- Topic: Discretization theoretical foundation
- Replies: 1
- Views: 2672
Re: Discretization theoretical foundation
You mean if there is any theoretical argument for optimality of uniform width or uniform counts? I am not aware of any general argument, although I would expect balancing the number of points in each of the intervals to perform better than having some intervals with few data points (unless there are...
- Thu Jun 08, 2017 8:24 pm
- Forum: GeNIe
- Topic: Equation Learning
- Replies: 6
- Views: 5248
Re: Equation Learning
Martin,
The planned release date is late June/early July. I will be happy to send you a link to our beta version. Will follow this up with a personal note.
Cheers,
Marek
The planned release date is late June/early July. I will be happy to send you a link to our beta version. Will follow this up with a personal note.
Cheers,
Marek
- Thu Jun 01, 2017 10:21 am
- Forum: GeNIe
- Topic: Equation Learning
- Replies: 6
- Views: 5248
Re: Equation Learning
Martin, A great direction for work, not necessarily specific to Bayesian networks but rather to all system modeling tools. We are modeling precisely to be able to understand the system. Bayesian networks are closely related to systems of simultaneous equations. We are about to release a version of G...
Re: MCMC
I'm afraid SMILE is not open source but you can read about the EPIS-BN algorithm in the following paper: Yuan, Changhe & Marek J. Druzdzel. An Importance Sampling Algorithm Based on Evidence Pre-propagation. In Proceedings of the Nineteenth Conference on Uncertainty in Artificial Intelligence (U...
- Mon May 15, 2017 8:22 pm
- Forum: GeNIe
- Topic: How do I get a conditional probability table?
- Replies: 1
- Views: 2923
Re: How do I get a conditional probability table?
I'm afraid no Bayesian network software will allow you to do that -- you have a theoretical problem. You need P(C|AB) and have only P(C|A) and P(C|B). Without special assumptions about the interaction, you are facing an under-constrained problem. Please look at the canonical models (Section "Co...