2013/7/30 Wifi Gi <[email protected]>:
> Oliver,
> I am well aware that joblib does the saving TO DISK, but if the model is not
> saved (or stored, if you prefer that term) in the Pipeline data structure,
> it doesn't help me to write that data structure to disk. My question was
> whether the estimator in the Pipeline is the fitted classifier or simply an
> instance of the class with no training.

It references an instance of the class with training, otherwise the
predict function of the Pipeline object would not be able to do any
reasonable predictions.

I agree that this is a bit inconsistent with other estimators where
attributes at fit time have a trailing `_` in the attribute name while
constructor parameters are left untouched.

-- 
Olivier

------------------------------------------------------------------------------
Get your SQL database under version control now!
Version control is standard for application code, but databases havent 
caught up. So what steps can you take to put your SQL databases under 
version control? Why should you start doing it? Read more to find out.
http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to