Partial Dependence and RandomForest

Please read the help page for the partialPlot() function and make sure you learn about all its arguments (in particular, "which.class").

Andy 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of jmc
Sent: Wednesday, April 11, 2012 2:44 PM
To: r-help at r-project.org
Subject: [R] Partial Dependence and RandomForest

Hello all~

I am interested in clarifying something more conceptual, so I won't be
providing any data or code here.

From what I understand, partial dependence plots can help you understand the

relative dependence on a variable, and the subsequent values of that
variable, after "averaging out the effects" of the other input variables. 
This is great, but what I am interested in knowing is how that relates to
each predictor class, not just the overall prediction.

Is it possible to plot partial dependence per class?  Specifically, I'd like
to know the important threshold values of my most important variables.

Thank you for your time,


--
View this message in context: http://r.789695.n4.nabble.com/Partial-Dependence-and-RandomForest-tp4549705p4549705.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Notice:  This e-mail message, together with any attachme...{{dropped:11}}

Thank you Andy.  I obviously neglected to read into the help file and,
frustratingly, could have known this all along.  However, I am still
interested in knowing the relative maximum value in the partial plots via
query instead of visual interpretation (and possibly getting at other
statistical measures like standard deviation).  Is it possible to do this? 
I will keep investigating, but would appreciate a hint in the right
direction if you have time.

--
View this message in context: http://r.789695.n4.nabble.com/Partial-Dependence-and-RandomForest-tp4549705p4555146.html
Sent from the R help mailing list archive at Nabble.com.

Note that the partialPlot() function also returns the x-y pairs being plotted, so you can work from there if you wish.  As to SD, my guess is you want some sort of confidence interval or band around the curve?  I do not know of any theory to produce that, but that may well just be my ignorance.

Andy 

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of jmc
Sent: Friday, April 13, 2012 11:20 AM
To: r-help at r-project.org
Subject: Re: [R] Partial Dependence and RandomForest

Thank you Andy.  I obviously neglected to read into the help file and,
frustratingly, could have known this all along.  However, I am still
interested in knowing the relative maximum value in the partial plots via
query instead of visual interpretation (and possibly getting at other
statistical measures like standard deviation).  Is it possible to do this? 
I will keep investigating, but would appreciate a hint in the right
direction if you have time.

--
View this message in context: http://r.789695.n4.nabble.com/Partial-Dependence-and-RandomForest-tp4549705p4555146.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Notice:  This e-mail message, together with any attachme...{{dropped:11}}

Hi

If I had a dataset of the following below and I want to plot colors for this image.plot so that:
red for data above 1
black for data between/equal 1 and -1
green for data below -1

normally its:
image.plot(data) but how do I add the colors for the range of data points as above?


Many thanks for your help!


_DATA_
2.04
2.02
1.60
1.92
0.98
0.50
0.40
-0.50
-0.40
-1.90
-2.8

Hi,

?image.plot(x, breaks=c(-100,-1,1,100), col=c('green','black','red'))

Regards,
Pascal


----- Mail original -----
De?: David Lyon <david_lyon3 at yahoo.com>
??: "r-help at r-project.org" <r-help at r-project.org>
Cc?: 
Envoy? le : Mercredi 18 avril 2012 0h00
Objet?: [R] image.plot specify colors for range of data points, please help

Hi

If I had a dataset of the following below and I want to plot colors for this image.plot so that:
red for data above 1
black for data between/equal 1 and -1
green for data below -1

normally its:
image.plot(data) but how do I add the colors for the range of data points as above?


Many thanks for your help!


_DATA_
2.04
2.02
1.60
1.92
0.98
0.50
0.40
-0.50
-0.40
-1.90
-2.8

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Thank you so much!




----- Original Message -----
From: Pascal Oettli <kridox at ymail.com>
To: David Lyon <david_lyon3 at yahoo.com>
Cc: "r-help at r-project.org" <r-help at r-project.org>
Sent: Tuesday, April 17, 2012 11:31 AM
Subject: Re : [R] image.plot specify colors for range of data points, please help

Hi,

?image.plot(x, breaks=c(-100,-1,1,100), col=c('green','black','red'))

Regards,
Pascal


----- Mail original -----
De?: David Lyon <david_lyon3 at yahoo.com>
??: "r-help at r-project.org" <r-help at r-project.org>
Cc?: 
Envoy? le : Mercredi 18 avril 2012 0h00
Objet?: [R] image.plot specify colors for range of data points, please help

Hi

If I had a dataset of the following below and I want to plot colors for this image.plot so that:
red for data above 1
black for data between/equal 1 and -1
green for data below -1

normally its:
image.plot(data) but how do I add the colors for the range of data points as above?


Many thanks for your help!


_DATA_
2.04
2.02
1.60
1.92
0.98
0.50
0.40
-0.50
-0.40
-1.90
-2.8

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.