Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[java-nlp-user] MAP estimate

Christopher Manning manning at stanford.edu
Fri May 6 13:23:29 PDT 2011


You could poke around what you can find in the optimization (and classify) packages.  I don't think we have anything that does coordinate ascent, but you can find other optimizers such as gradient ascent and conjugate gradient and implementations of gaussian priors.  But the internals aren't all very well documented, so if you don't have a reasonable idea what you're doing, something else might be easier to work with....

Chris.


On May 5, 2011, at 11:47 PM, Suriya Gunasekar wrote:

> I am using stanford NLP for the first time. I need some class which can find a MAP(maximum a-posteriori) estimate of a random variable with gaussian prior using coordinate ascent or em alogrithms. it will be of great help if someone can help me with this.
> 
> suriya
> _______________________________________________
> java-nlp-user mailing list
> java-nlp-user at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/java-nlp-user




More information about the java-nlp-user mailing list