[SciPy-dev] MCMC, Kalman Filtering, AI for SciPy?

[Robert Kern]

chris at fisher.forestry.uga.edu wrote:
> For the past couple of years, I have been maintaining a package called
> PyMC (http://pymc.sourceforge.net) which provides a general Markov chain
> Monte Carlo sampler (Metropolis-Hastings), linear and non-linear Kalman
> filtering algorithms and several reinforcement learning algorithms for
> AI applications. To date, it has been used mostly for ecological
> modeling applications, but is far more general than that. This package
> currently relies on SciPy, particularly for such tasks as plotting, and
> uses f2py extensively.
> 
> I am wondering if there is interest in integrating such functionality
> into SciPy. If so, I would be happy to contribute, integrate and
> maintain the code under the auspices of the SciPy project. If not, I am
> equally content to continue maintaining PyMC separately. I encourage
> those interested to check out PyMC and let me know if it would be
> worthwhile to add any of this to SciPy.

Fantastic!

I am definitely interested in seeing MCMC and related algorithms 
integrated into SciPy. I didn't look too hard at at your code when you 
first announced it because I wanted something to go into SciPy even if I 
had to write it myself, and we're not incorporating LGPL code into 
SciPy. But if you're willing to relicense it under the BSD license, I 
don't have to redo all the hard work you've done. And that makes me happy.

It would be worthwhile to go make sure the MCMC design is general enough 
to accommodate other sampling strategies (specifically, I'd like to 
implement some of the ideas of Radford M. Neal[1]). Gibbs sampling from 
models specified in a manner similar to BUGS would be cool, too.

[1] http://www.cs.toronto.edu/~radford/fbm.software.html

-- 
Robert Kern
rkern at ucsd.edu

"In the fields of hell where the grass grows high
  Are the graves of dreams allowed to die."
   -- Richard Harter