Gradient-Boosted Trees (GBTs) are a type of ensemble classification algorithm that uses decision trees to build a predictive model. GBTs can take numeric or categorical input variables and can classify a binary target variable or predict a numeric target with regression.
Learn more about this implementation from the MLlib Documentation
- SPSS Modeler v18.0 or later
- Python 2.7 Anaconda Distribution
More information here: IBM Predictive Extensions
If using v18.0 of SPSS Modeler, navigate to the options.cfg file (Windows default path: C:\Program Files\IBM\SPSS\Modeler\18.0\config). Open this file in a text editor and paste the following text at the bottom of the document:
eas_pyspark_python_path, "C:/Users/IBM_ADMIN/Anaconda/python.exe"
- The italicized path should be replaced with the path to your python.exe from your Anaconda installation.
- Go to the Extension menu in Modeler and click "Extension Hub"
- In the search bar, type the name of this extension and press enter
- Check the box next to "Get extension" and click OK at the bottom of the screen
- The extension will install and a pop-up will show what palette it was installed to
- Save the .mpe file to your computer
- In Modeler, click the Extensions menu, then click Install Local Extension Bundle
- Navigate to where the .mpe was saved and click open
- The extension will install and a pop-up will show what palette it was installed
- Nial McCarrol - (www.mccaroll.net)
- Greg Filla (gdfilla)