2014-09-23 23:05 GMT+02:00 Pagliari, Roberto :
> I didn’t realize that. I thought it was the name of the class.
It is when you use the make_pipeline shorthand.
--
Meet PCI DSS 3.0 Compliance Requirements with EventLog Ana
I didn’t realize that. I thought it was the name of the class.
Thank you,
From: Eraldo Pomponi [mailto:[email protected]]
Sent: Tuesday, September 23, 2014 4:56 PM
To: [email protected]
Subject: Re: [Scikit-learn-general] name of random forest classifier params
Dear Roberto,
On Wed, Sep 24, 2014 at 12:14 AM, Pagliari, Roberto wrote:
> I’m using a pipeline with gridsearchcv. I tried this to allow search over
> a range of number of trees
>
>
>
> params = dict(random_forest_classifier__n_estimators=[8, 9, 10, 11])
>
> clf = grid_search.GridSearchCV(my_pip
I'm using a pipeline with gridsearchcv. I tried this to allow search over a
range of number of trees
params = dict(random_forest_classifier__n_estimators=[8, 9, 10, 11])
clf = grid_search.GridSearchCV(my_pipeline, param_grid=params)
but the name 'random_forest_classifier__n_estimators' is not co
