On Tue, Feb 28, 2012 at 8:21 AM, Olivier Grisel
<[email protected]> wrote:
> 2012/2/28 Brent Pedersen <[email protected]>:
>> Hi,
>> is there an example in the docs that I'm missing of using
>> PCA to remove batch effects?
>
> I don't really know what you mean by batch effects but I assumes you
> want to get rid of low frequency components in images or sounds or
> noisy generic topics (such as generic stop words) in text documents.
>

In this case, I am looking at microarray data. Batches done on
different dates or in different labs tend to cluster together.

>> If I understand correctly, I may be able to do this by
>> using U, S, V = pca._fit(X) and then removing the first n_components.
>> Is there a way to do this so I can then use inverse_transform
>> to get the data in the original projection with batch effects
>> removed?
>
> I think you should write your own class that derives from sklearn PCA
> or RandomizedPCA and implement transform and inverse_transform as you
> wish.
>

I will look into this. Thanks.


> --
> Olivier
> http://twitter.com/ogrisel - http://github.com/ogrisel
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to