[SciPy-dev] BayesNet SciKit

[Ed Rahn]

What's the reasoning for starting from scratch on a port instead of 
using preexisting code that has the same exact objective?

- Ed

Gouws, Almero <14366037 at sun.ac.za> wrote:
> Hi,
> 
> I am a masters student currently working on a port of Kevin P. Murphy's
> Matlab toolbox, FullBNT, to Python. The code is not yet available in a
> public repository, because I would like it to match or surpass 
> the functionality of the openbayes toolbox before making it available.
> 
> Here's a list of what is currently working:
> 
> -Static Bayesian Networks:
>  -Nodes:
>   - Mutinomial nodes:
>    - (Sequential) Bayesian parameter updating from complete data.
>    - Maximum liklihood parameter estimation from complete data.
>    - Maximum likelihood parameter estimation with missing values (EM)
>    - Greedy K2 structure learning.
> 
>   - Gaussian nodes:
>    - Maximum liklihood parameter estimation from complete data.
> 
>  - Inference Engines:
>   - Junction tree inference engine
>     - Works for Multinomial and Guassian nodes.
> 
> Thanks
> Almero Gouws
> 
> _______________________________________________
> Scipy-dev mailing list
> Scipy-dev at scipy.org
> http://projects.scipy.org/mailman/listinfo/scipy-dev
>