You have it right, I believe.
As far as jSMILE goes, it is open source and you can easily add a wrapper around the SMILE function that you want exposes to Java.
Cheers,
Marek
Search found 430 matches
- Mon Jan 23, 2017 11:02 pm
- Forum: SMILE
- Topic: Setting Beta(Dirichlet) prior when using jsmile EM
- Replies: 4
- Views: 5245
- Mon Jan 23, 2017 10:48 pm
- Forum: GeNIe
- Topic: The effect of fixed nodes in EM
- Replies: 8
- Views: 6304
Re: The effect of fixed nodes in EM
Hi Timo, I'm having a hard time understanding your questions but let me try. Please let me know in case I have misunderstood you. I'm not sure what you mean by setting nodes as evidence and simulating scenarios during learning parameters. Have you kept refining the same set of parameters with succes...
- Thu Jan 12, 2017 7:23 pm
- Forum: SMILE
- Topic: Setting Beta(Dirichlet) prior when using jsmile EM
- Replies: 4
- Views: 5245
Re: Setting Beta(Dirichlet) prior when using jsmile EM
I'm afraid the only parameter that you can specify in learning parameters is confidence, which is the equivalent sample size (ESS). This ESS will be for all parameters in the entire network, so it is not possible to specify different ESS for different parameters. I guess you would want at most one E...
- Thu Jan 05, 2017 10:38 am
- Forum: GeNIe
- Topic: Naive Bayes
- Replies: 13
- Views: 9994
Re: Naive Bayes
Glad to be of service -- Marek
- Thu Jan 05, 2017 10:34 am
- Forum: SMILE
- Topic: smile Tutorials
- Replies: 5
- Views: 5409
Re: smile Tutorials
Have you tried right-click/"Save As" :-)? -- Marek
- Wed Jan 04, 2017 9:35 pm
- Forum: GeNIe
- Topic: The effect of fixed nodes in EM
- Replies: 8
- Views: 6304
Re: The effect of fixed nodes in EM
When you fix them, whether they are present or absent in the data file, does not matter, as their CPTs will remain unchanged. If you do not fix them and they are absent from the data file, EM will calculate their CPTs in every iteration based on the values of other variables in the data file. The CP...
Re: CPT计算
Gary, I'm not sure I understand your question if I take it for the face value. Each column in a CPT is in general independent of any other column, so your question does not have a good answer in general. However, your question may be interpreted differently: "Is there any structure in the CPT t...
- Thu Dec 22, 2016 9:10 am
- Forum: GeNIe
- Topic: The effect of fixed nodes in EM
- Replies: 8
- Views: 6304
Re: The effect of fixed nodes in EM
Timo, You may be reading too much into fixing nodes. What fixing nodes does is excluding them from the learning process. In other words, their CPTs do not get affected by the learning. Try fixing some nodes and not fixing others. If you do this, the nodes that you have fixed will preserve their orig...
- Fri Dec 09, 2016 1:39 pm
- Forum: GeNIe
- Topic: Does GeNIe support dynamic bayesian decision networks (dynamic influence diagrams)?
- Replies: 3
- Views: 3917
Re: Does GeNIe support dynamic bayesian decision networks (dynamic influence diagrams)?
Please stay tuned -- it is one of the functionalities on our TO DO list.
Cheers,
Marek
Cheers,
Marek
- Mon Dec 05, 2016 1:50 pm
- Forum: GeNIe
- Topic: Naive Bayes
- Replies: 13
- Views: 9994
Re: Naive Bayes
Got your data. It has only 38 data records (very few) and 23 variables. What is it exactly that you would like to do with the data? Replicating the results from the paper will be hard, if not impossible, given that you don't know which of the records were used for training and which for testing. Was...
- Mon Nov 28, 2016 7:47 am
- Forum: GeNIe
- Topic: Naive Bayes
- Replies: 13
- Views: 9994
Re: Naive Bayes
Your first question cannot be answered without data. I looked at the paper but cannot find the source of the data. Do you have the data or know where I can take it from? We can see whether variables are independent in the data or not. log(p) is the overall log likelihood score of the final iteration...
- Tue Nov 15, 2016 11:15 pm
- Forum: GeNIe
- Topic: MAX doesn't work
- Replies: 9
- Views: 7269
Re: MAX doesn't work
I have examined your model and I believe that the function is working fine. What it does is the following: For every sample generated for the nodes C2 and C3, it picks the higher of the two values as the sample for C23. What you see with the default number of samples (1000) is the distribution of sa...
- Tue Nov 15, 2016 12:10 pm
- Forum: GeNIe
- Topic: MAX doesn't work
- Replies: 9
- Views: 7269
Re: MAX doesn't work
I'm not sure why you are saying that MAX does not work. I have just created a network as you described and tested it. I have run the standard number of samples and the child node (with the MAX function) seems to generate a correct distribution. I have changed the number of samples to 1 and the resul...
- Wed Nov 09, 2016 7:31 am
- Forum: GeNIe
- Topic: Naive Bayes
- Replies: 13
- Views: 9994
Re: Naive Bayes
At least one of the columns in your data file contains a constant value, as you say. Constant values are useless in learning. The common sense of this is the following: If a player is present during each of the games, then his/her presence cannot be a predictor whether the team wins. He/she will be ...
- Mon Nov 07, 2016 9:30 pm
- Forum: GeNIe
- Topic: Naive Bayes
- Replies: 13
- Views: 9994
Re: Naive Bayes
I'm not sure what you are doing, so I cannot give you a 100% certain answer. If you have the same data set and the same model and perform Validation multiple times, you can get different results each time if you use Folding seed of 0. In that case, GeNIe will use the system clock to start the random...