BayesFusion Support Forum

self.change_evidence_and_update(net, 'e_proponer', 1, 'Bajo') self.change_evidence_and_update(net, 'e_aportar', 2, 'Bajo') self.change_evidence_and_update(net, 'e_proponer', 4, 'Medio') self.change_evidence_and_update(net, 'e_aportar', 5, 'Medio') self.change_evidence_and_update(net, 'e_colaborar',...

We have noticed that something is not quite right with network definition. You probably have not set nodes definition for t=0 (ex. Proponer_E) def_ex.PNG It should be set with method set_node_definition (not like other slices - set_node_temporal_definition). Also notice, that in SMILE 'time' variabl...

def print_posteriors(net, node_handle): node_id = net.get_node_id(node_handle) if net.is_evidence(node_handle): print(node_id + " has evidence set (" + net.get_outcome_id(node_handle, net.get_evidence(node_handle)) + ")") else : posteriors = net.get_node_value(node_handle) n = n...

Why beliefs initialization goes through 10 steps in time? It depends on Slice (Step) Count variable. You can change it by calling ex. net.set_slice_count(4) (four steps). Remember to update beliefs. How do I add evidences along time? (I have a guess on this one, but I ask it just in case I am wrong...

def print_posteriors(self, net, node_handle): node_id = net.get_node_id(node_handle) if net.is_evidence(node_handle): print(node_id + " has evidence set (" + net.get_outcome_id(node_handle, net.get_evidence(node_handle)) + ")") else : posteriors = net.get_node_value(node_handle)...

I'll try to explain PySmile DBN usage based on GeNIe documentation: https://support.bayesfusion.com/docs/GeNIe/dbns_creatingdbn.html 1. Create CPT nodes, ex. net.add_node("Location") net.add_node("Rain") net.add_node("Umbrella") 2. Set node temporal types plate_explaine...

1. I'm assuming, that you initialized dataset and network properly (as ex. ds and net). node_id = "someNodeIdentifier" matching = ds.match_network(net) validator = pysmile.Validator(net, ds, matching) validator.add_class_node(node_id) em = pysmile.Learning.EM() em.learn(ds, net, matching) ...

As you said - it's mismatch between unicode used to Python compilation. Unfortunately, we don't have build dedicated for Python builded with UCS2. At this moment, only way to use PySmile is to have Python compiled with UCS4.

Hello,

I need some informations first, if you want me to help you.

1. Which OS you have?
2. Which Python you are using? (python -V)
3. Your Python Unicode config (print sys.maxunicode)

Best regards,
Piotr

Unfortunately, the only way you can do it is writing validation yourself following standards of cross validation. We have validation implemented for 'hard' evidence only.

It seems that your network has non-fixed deterministic and Noisy Adder nodes (which are not supported by EM algorithm). You should try other learning algorithm or modify your network to fit EM.

I tried to reproduce that error with random data, but it was ineffective. I suspect the end of RAM memory, but 24x400 DataSet is not that big. If your data is not confidential, archive your project and data into ZIP or RAR and send as attachment to support@bayesfusion.com We will do our best to help...

I forgot to ask about JVM error stack - it could be helpful.

Also, did you matched BKK with DataSet?
(code should be executed after bkk initialization)

Last thing - did forced and forbidden arcs aren't excluding each other?

Hello,

Could you show part of code causing that error?

I'm assuming, that your Dataset isn't corresponding with Network.
Net_tut_5.xdsl has states named State0/State1 instead of t/f from Dataset.