OOBPredictorImportance How to interpret the values range?

1 次查看（过去 30 天）
Tiago Dias
Tiago Dias 2020-1-6
评论： Tiago Dias 2020-6-4
Hello,
To be able to calculate the importance of the predictors for TreeBagger it is needed to turn on the OOBPredictorImportance parameter
Mdl = TreeBagger(200,X,'MPG','Method','regression','Surrogate','on','PredictorSelection','curvature','OOBPredictorImportance','on');
I would like to know what is the range of the parameters, the results for my regression is beetween 0 and 1.5, is that good? bad? What more conclusions can I draw from that vector?
Thanks for your time!
Tiago

请先登录，再回答此问题。

回答（1 个）

Adrian Jacinto
Adrian Jacinto 2020-6-4
Hi, really this parameter have no range. This parameter only explain you how important is a band compared to other bands. When you a have a higuest values associated with a band, this indicates that this band is more important than the rest.
You can make the result more visual by applying the bar () function on the OOB vector to make a bar diagram.
  1 个评论
Tiago Dias
Tiago Dias 2020-6-4
Ok, thanks for you input.
I tought that perhaps it would be on percentage or something like that. IT is more quanlitative then quantitative.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Classification Ensembles 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by