Hi Luca,
The "best" strategy consists in finding the best threshold, that is the one
that maximizes impurity decrease, when trying to partition a node into a
left and right nodes. By contrast, "random" does not look for the best
split and simply draw the discretization threshold at random.
For further details, you can have a look at section 3.6.3 of my thesis. It
describes the algorithm as it is implemented in Scikit-Learn.
http://www.montefiore.ulg.ac.be/~glouppe/pdf/phd-thesis.pdf
Hope this helps,
Gilles
On 12 September 2014 16:11, Luca Puggini <[email protected]> wrote:
> Hi,
> I am using random forest classifier and this algorithm train a tree
> defined as :
>
> DecisionTreeClassifier(criterion='gini', max_depth=None,
> max_features='auto',
> max_leaf_nodes=None, min_samples_leaf=1, min_samples_split=2,
> min_weight_fraction_leaf=0.0, random_state=1982007276,
> splitter='best')]
>
> I do not understand what algorithm is used to train a tree with this
> parameters.
> Is there any reference that describes the used training algorithm in
> details?
>
> In particular I do not understand the split strategy
>
> splitter : string, optional (default="best")
> The strategy used to choose the split at each node. Supported strategies
> are "best" to choose the best split and "random" to choose the best
> random split.
>
>
> Thanks for help,
> Luca
>
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
>
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
>
------------------------------------------------------------------------------
Want excitement?
Manually upgrade your production database.
When you want reliability, choose Perforce
Perforce version control. Predictably reliable.
http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
