BayesFusion Support Forum

That's great! I need to check this out.

Hi,

Thank you for your response. That would be absolutely great. But even easier, it would be totally sufficient to provide a method which reads XDSL file from an array of raw bytes instead of a file path. This way the user's application could read from any source, store and preprocess data in the ...

Hi,

Just my two cents It would be great to have a SMILE function which could get the data from a stream instead of a file. You can always open a file as a stream but you cannot load your data from a custom source (i.e. encoded XDSL file stored on disk / in memory / database).

Kind Regards,
Piotr

Great, thanks

Hello,

Is it possible to generate data file with only target / observation nodes and skipping auxiliary nodes?

Thanks,
Piotr

Thanks for the advice, it's building now. I was wrong about the other conflicting changes, I must have missed something before. Still, waiting for the next release.

Regards

Hi,

when trying to compile jSMILE from available sources (http://genie.sis.pitt.edu/download/jsmile_src.zip, December 19, 2011) with latest SMILE (http://genie.sis.pitt.edu/download/smile_1_1_windows_vc10.zip, April 24, 2012) I get following error (Win32, both Debug and Release, VS2010):


1>smile ...

Thanks that's great news!

Ok Remember that if you use jSmile the memory is managed by virtual machine, thus you would need to manually increase heap space.

Anyways, have you considered out of memory exceptions? Unfortunately it's not easy to check if SMILE runs out of memory.

According to the last post here: http://genie.sis.pitt.edu/forum/viewtopic.php?f=2&t=928
you cannot DeactivateRelevance() from GeNIe.

Cheers

Thanks for the update!

Hi,

I double your suggestion. I'd be great if you could load an XDSL model from a stream or an array of bytes. I think it's not that big deal and it would enable developers to 'hide' their models in compiled applications. Also it would provide a mean of encryption and decryption of models (without ...

Thank you Marek, I'll contact you via e-mail.

Firstly, sorry. I didn't inted to offend you (or anyone involved in SMILE) nor I wanted to sound rude. I hope that my poor language can be a bit of explanation. I really appreciate your work and I am a huge fan of SMILE.

Want I wanted to say is that I really believe that SMILE could develop faster ...