Chronological Thread 

< Chronological >      < Thread >

---

Hi Jason,

Thanks for the quick response! I too will be out on vacation next week. I
haven't tried categorical predictors yet, but am planning on it at some point
in the future. I'll give it a shot and let you know how it goes.

-Tom
------------------------------------------------------------------
Thomas Dilts, MS
Research Scientist
Great Basin Landscape Ecology Lab
Department of Natural Resources and Environmental Science
University of Nevada Reno
(775) 784-1447
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.cabnr.unr.edu_weisberg_&d=CwIFAg&c=imBPVzF25OnBgGmVOlcsiEgHoG1i6YHLR0Sj_gZ4adc&r=xCrbnFZcjrfVEdKQhMVPxwdYw2imyfElky5AJEHT_VA&m=92zHlnGhUPeFdUxoT3ul30CxrAqCGVIN4LAVYQN1Fbk&s=mm9KVQWT_ZhXc4alK6bXOFJNA-BwFr-UFZGjbqynaj0&e=

https://urldefense.proofpoint.com/v2/url?u=http-3A__gislandscapeecology.blogspot.com_&d=CwIFAg&c=imBPVzF25OnBgGmVOlcsiEgHoG1i6YHLR0Sj_gZ4adc&r=xCrbnFZcjrfVEdKQhMVPxwdYw2imyfElky5AJEHT_VA&m=92zHlnGhUPeFdUxoT3ul30CxrAqCGVIN4LAVYQN1Fbk&s=gIOFU9KDx7ZKFuMdVtYHV9leZQBp5VS7IyKKffN_PP8&e=


________________________________________
From: Jason Roberts []
Sent: Friday, July 01, 2016 3:12 PM
To: Dilts, Thomas
Cc:
Subject: RE: labelling categorical predictors in RandomForest

Tom,

Thanks for positive feedback!

For the RF question, I presume you are referring to the partial dependence
plots that can be optionally outputted. I haven't used the tool with
categorical predictors in quite a while so can't remember what it does in
that case. My memory is that when a categorical predictor is plotted, it
produces a bar for each categorical value. Is that what you mean, and is that
not happening? Anyway, sorry for my confusion. If you could describe a bit
more explicitly what you're doing, I can try to help. (You can send example
data, if that helps demonstrate what you're seeing.)

Regarding the forum, this is it. We are a bit old-school here and are still
using our listsrv. But I answer 98% of the questions, so you're not likely to
hear from anyone else. :-)

I'm on vacation next week so if you reply I will not get back to you until
July 11.

All the best,

Jason

-----Original Message-----
From:
[] On Behalf Of Dilts, Thomas
Sent: Thursday, June 30, 2016 5:39 PM
To:
Subject: [mget-help] labelling categorical predictors in RandomForest

Hi,

First off this is one of the best ArcGIS extensions of all time! Thanks for
the great work on developing it.

Q: I was wondering if there is a way to label the categorical predictors so
that it is possible to identify 1=shrub, 2=forest, 3=wetland, etc. on the
output response curves. For example Maxent software requires each variable
to be identified as continuous or categorical. Is there a similar option
with RF?

Q: Is there a users group of some sort of forum to ask questions on?

Thanks

Sincerely,

Tom Dilts

------------------------------------------------------------------
Thomas Dilts, MS
Research Scientist
Great Basin Landscape Ecology Lab
Department of Natural Resources and Environmental Science University of
Nevada Reno
(775) 784-1447
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.cabnr.unr.edu_weisberg_&d=CwIFAw&c=imBPVzF25OnBgGmVOlcsiEgHoG1i6YHLR0Sj_gZ4adc&r=xCrbnFZcjrfVEdKQhMVPxwdYw2imyfElky5AJEHT_VA&m=j4IpST20f1kqbaG5sPe3CbFj-8oDUhvttPIzYpP0sf8&s=IgaUELP9s5HPKLUQk_xZShaqNDeBdQZpGHhSZUItnxY&e=
https://urldefense.proofpoint.com/v2/url?u=http-3A__gislandscapeecology.blogspot.com_&d=CwIFAw&c=imBPVzF25OnBgGmVOlcsiEgHoG1i6YHLR0Sj_gZ4adc&r=xCrbnFZcjrfVEdKQhMVPxwdYw2imyfElky5AJEHT_VA&m=j4IpST20f1kqbaG5sPe3CbFj-8oDUhvttPIzYpP0sf8&s=iK-7QWSbHOM4euEFBvXcqWbB8OLUbu6y4wOCxFIOLkk&e=

---

• RE: [mget-help] labelling categorical predictors in RandomForest, Jason Roberts, 07/01/2016
  • RE: [mget-help] labelling categorical predictors in RandomForest, Dilts, Thomas, 07/01/2016

---

Archive powered by MHonArc 2.6.19.