Search found 4 matches

by korfan01
Tue Oct 27, 2015 7:38 am
Forum: GeNIe
Topic: Virtual Evidence for Validation/Testing
Replies: 1
Views: 3217

Virtual Evidence for Validation/Testing

Is it possible to add virtual evidence to a variable by assigning this evidence into a test dataset and give this as input at the validation or testing process? Also is it possible to learn the parameters by assigning probability distributions as values to the variables in the training dataset input...
by korfan01
Wed Oct 08, 2014 4:31 am
Forum: GeNIe
Topic: DBN Learning parameters in a non unrolled network
Replies: 0
Views: 16221

DBN Learning parameters in a non unrolled network

What is the difference in learning the parameters for DBN in an unrolled and in a non-unrolled network? Does the process of learning the parameters in a non-unrolled network use the records from all time slices included in the dataset, and if so how are they combined? For example assuming three time...
by korfan01
Thu Jun 05, 2014 6:35 am
Forum: GeNIe
Topic: Validation methods in GeNIe
Replies: 5
Views: 6401

Re: Validation methods in GeNIe

One last question, how can I use cross validation to evaluate a DBN prognostic model? Let the DBN has 5 variables: [A,B,C,D,E] and it is unrolled for 3 time slices [t0, t1, t2] and the goal is to predict the class value at t=2 (last time slice) e.g. E_2, given all the observations up to t=1 [A_0,A_1...
by korfan01
Fri Jan 31, 2014 8:58 pm
Forum: GeNIe
Topic: Validation methods in GeNIe
Replies: 5
Views: 6401

Validation methods in GeNIe

Hi, I used validation for DBN in Genie but I am not sure how the output prediction probabilities are computed. For example, is leave-one-out method repeated N times (where N is the data size) using N-1 records for training and leave one record for testing (leaving a different record out on each repe...