ok, this is what I can gather from the code:
================================
Expectation Step
--------------------------
Calculate the loglikelihood and responsibilities for each sample.
a. for each sample the loglikelihood is calculated for each gaussian
and then sum across models (logprob = logsumexp(lpr, axis=1))
b. responsibilites is just lpr, but normalized to 1 across each row?
This is figure out which gaussian model is responsible for the
loglikelihood of this sample correct?
Maximization Step
---------------------------
Need help here, can't quite figure out whats going on
================================
As you can see the maximization step is where things fall apart for me.
I've followed many tutorials, including this (
https://www.youtube.com/watch?v=AnbiNaVp3eQ). They eventually involve
sufficient statistics, etc. which seem to make some sense at some high
level, but I can never connect that theory to code, in particular the code
in the GMM module.
Can someone help me fill in the blanks? Sorry if this is a little open
ended
-Dave
On Tue, Sep 3, 2013 at 4:51 PM, Olivier Grisel <[email protected]>wrote:
> 2013/9/3 David Reed <[email protected]>:
> > Thanks Jake, I will look into this more. Is this the proper forum to
> ask
> > questions about its implementation if I am confused?
>
> Yes it is.
>
> --
> Olivier
> http://twitter.com/ogrisel - http://github.com/ogrisel
>
>
> ------------------------------------------------------------------------------
> Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
> Discover the easy way to master current and previous Microsoft technologies
> and advance your career. Get an incredible 1,500+ hours of step-by-step
> tutorial videos with LearnDevNow. Subscribe today and save!
> http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
