cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Check out the JMP® Marketplace featured Capability Explorer add-in
Choose Language Hide Translation Bar
uejimurashu
uejimurashu
Level I

Is boosted tree in JMP the same as gradient boosted tree?

Dec 28, 2023 01:29 AM (1068 views)

Is boosted tree in JMP the same as gradient boosted tree?
In many academic papers, the term gradient boosted tree is used, and I could not find the word boosted tree.

0 Kudos
1 ACCEPTED SOLUTION

Accepted Solutions
SDF1
Super User SDF1
Super User
Solution

Re: Is boosted tree in JMP the same as gradient boosted tree?

Dec 28, 2023 08:40 AM (1051 views)  |  Posted in reply to message from uejimurashu 12-28-2023

Hi @uejimurashu ,

 

  Welcome to the JMP Community pages. In short, the answer is no, they are not the same. JMP has several different tree-based platforms that all have slightly different algorithms and implementations to how the modeling is optimized, as well as the hyperparameters for the model. These include partitioning, boosted trees, bootstrap forest (similar to random forest), and extreme gradient boosted trees (XGBoost). The last one is an additional add-in, that I believe you need at least JMP Pro v16.0 or higher to run -- it could be v15.0.

 

  You can find out more in JMP Help about boosted trees here, bootstrap forests here, and partitioning here. There's a good video on the different platforms here, and if you have Pro, then you can read up on XGBoost here.

 

Hope this helps!,

DS

View solution in original post

1 REPLY 1
SDF1
Super User SDF1
Super User
Solution

Re: Is boosted tree in JMP the same as gradient boosted tree?

Dec 28, 2023 08:40 AM (1052 views)  |  Posted in reply to message from uejimurashu 12-28-2023

Hi @uejimurashu ,

 

  Welcome to the JMP Community pages. In short, the answer is no, they are not the same. JMP has several different tree-based platforms that all have slightly different algorithms and implementations to how the modeling is optimized, as well as the hyperparameters for the model. These include partitioning, boosted trees, bootstrap forest (similar to random forest), and extreme gradient boosted trees (XGBoost). The last one is an additional add-in, that I believe you need at least JMP Pro v16.0 or higher to run -- it could be v15.0.

 

  You can find out more in JMP Help about boosted trees here, bootstrap forests here, and partitioning here. There's a good video on the different platforms here, and if you have Pro, then you can read up on XGBoost here.

 

Hope this helps!,

DS