Can pre-trained machine learning models be imported into GENIE?

[Andy123]

Hello all!

I hope all is well with you. I am a new user to GENIE/SMILE and had a few questions I wanted to ask.

Is it possible to import a trained machine learning model, e.g. a feedforward neural network (FFNN), into GENIE? For example, is it possible to change the equation in an equation node from equation 1 to equation 2 below?

1) OutputVar1 = InputVar1 + InputVar2 + InputVar3 + Normal(0,std)
2) OutputVar1 = FFNN(InputVar1,InputVar2,InputVar3) + Normal(0,std)

Another question I have is if it is possible to have the FFNN set evidence from within GENIE? For example, the output of the FFNN can be a vector of fault probabilities of a machine and this information can then be used to set evidence in another part of the network. Is this something that is possible in GENIE?

I understand that this is much easier with SMILE and is the reason why the wrappers exist, but wanted to ask if integrating pre-trained machine learning models is possible so that interactions between the Bayesian network and the user can be through GENIE instead of the terminal.

Thank you so much for your time and help!

Andy

[marek [BayesFusion]]

Hi Andy,

I'm afraid you cannot do what you want automatically through GeNIe. The only way to achieve what you want is through SMILE, which is a general purpose library that can be used from inside your computer programs. Doing what you want should not be too hard using SMILE.

I hope this helps,

Marek

[Andy123]

Hi Marek,

Thank you so much for responding so quickly. Yes that explanation is helpful. I had one more question I wanted to ask. Is it possible to save evidence to a model? I noticed that whenever I set evidence in GENIE, save and close the model, and open it back up the node is reset back to not having any evidence. Is there a way to set the evidence once and for it to remain? Thank you again for your help, I really appreciate it.

Best,

Andy

[marek [BayesFusion]]

Hi Andy,

This is possible through cases. If you create or open a case library, you will be able to save the current states of evidence to a case. Cases are saved with the model, so you can load the case back once you have opened the model. You can find more information in GeNIe manual.
Cheers,

Marek

[Andy123]

Hi Marek,

Thank you so much for your help, this is exactly what I need. Thank you!

Best,

Andy