On 11/17/2012 04:19 PM, Ronnie Ghose wrote:
> See you guys just said I could use trees on subsets and they will work 
> well.
>
> So why not partial_fits + trees?
>
As I tried to say, these are different stories:
Gilles said (and wrote about) using a different small subset of the data 
for each tree.
In total, you still use all the data. You just build multiple models, 
each with a different
subset. So you never have to store all the data at once.
In the end you get many classifiers that you can bag.

The partial fit is something different. It builds a single model 
incrementally from the
data.
This is feasible only for some models and some ways to train models.
It is implemented for SGDClassifier mainly (I think).
It would be easy to implement this for the naive Bayes models, but
they don't have partial fit yet afaik.

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to