[SciPy-user] Bayes net question

[Karl Young]

David Warde-Farley wrote:

>On 16-Jan-08, at 12:05 PM, Karl Young wrote:
>
>  
>
>>I've looked at the documentation for that package (but refuse to go  
>>back
>>to Matlab re. actually using it :-)) and it looks like it's quite a  
>>bit
>>further along than OpenBayes re. options so does it seem like a
>>consensus that we should wait for this to be settled before any
>>decisions about incorporating Bayes nets ? As I mentioned before I'm
>>willing to help (e.g. with a port) once a decision is reached re. the
>>best approach.
>>    
>>
>
>Likewise, I'd be willing to help with a port.
>
>I know that BNT includes code from other projects (for one thing the  
>IRLS implementation for optimizing the weights of a softmax from the  
>NetLab neural networks toolbox), so it may not be totally up to  
>Professor Murphy whether or not to dual-license the entire thing.
>  
>

Good point though I guess we could study which of that functionality 
could be replaced by existing Scipy bits and which might be reasonable 
to rewrite (particularly if there are more than one of us working on 
this), e.g. we might not need everything that's in the neural networks 
toolbox.

-- 

Karl Young
Center for Imaging of Neurodegenerative Diseases, UCSF          
VA Medical Center (114M)              Phone:  (415) 221-4810 x3114  lab        
4150 Clement Street                   FAX:    (415) 668-2864
San Francisco, CA 94121               Email:  karl young at ucsf edu