View unanswered posts | View active topics
Forum rules
1. This forum can be used for discussion of both ALGLIB-related and general numerical analysis questions
2. This forum is English-only - postings in other languages will be removed.
 
|
Page 1 of 1
|
[ 1 post ] |
|
| Author |
Message |
|
msci
|
Post subject: Tuning of ratio r in alglib  Posted: Mon Aug 28, 2023 8:16 pm |
|
Joined: Mon Aug 28, 2023 8:11 pm
Posts: 1
|
|
I am tuning the parameter r (size of training set that is used in tree construction) in a random forest model (this is used to predict distance in km). I am evaluating both the OOB RMSE as well as the ratio RMSE Train / RMSE Test which should be close to 1.
r = 0.10 is giving OOB RMSE approx. 0.27 and ratio approx 0.9.
r = 0.15 is giving OOB RMSE approx. 0.26 and ratio approx. 0.85.
r = 0.2 is giving OOB RMSE approx. 0.25 and ratio approx. 0.8.
My question is: how close should the ratio of RMSE Train to RMSE Test be to 1, to be considered 'not much smaller than 1'?
|
|
| Top |
|
 |
|
|
Page 1 of 1
|
[ 1 post ] |
|
Who is online |
Users browsing this forum: No registered users and 6 guests |
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum
|
Login
Register
FAQ
Search
