cancel
Showing results for 
Search instead for 
Did you mean: 

How can I generate the decision tree? Is it possible to generate it with external tools? Which are the tools supported?

Evelina CAMILLERI
ST Employee

[webinar - Program decision trees in sensors with a Machine Learning Core]

1 ACCEPTED SOLUTION

Accepted Solutions
Evelina CAMILLERI
ST Employee

Unico GUI (from v9.8) has a built-in decision tree generation algorithm (developed by ST). However, other external tools for decision tree generation are supported as well, until the decision tree format is compatible with the decision tree format described in the MLC application note. 

In particular we have tested and supported Weka, RapidMiner, MATLAB and Python as external tools. The decision tree format generated by Weka is compatible with the Unico GUI, while for MATLAB and Python we provide some scripts (https://github.com/STMicroelectronics/STMems_Machine_Learning_Core/tree/master/tools) to be used as example to generate the decision tree in a textual format compatible with Unico GUI.

View solution in original post

1 REPLY 1
Evelina CAMILLERI
ST Employee

Unico GUI (from v9.8) has a built-in decision tree generation algorithm (developed by ST). However, other external tools for decision tree generation are supported as well, until the decision tree format is compatible with the decision tree format described in the MLC application note. 

In particular we have tested and supported Weka, RapidMiner, MATLAB and Python as external tools. The decision tree format generated by Weka is compatible with the Unico GUI, while for MATLAB and Python we provide some scripts (https://github.com/STMicroelectronics/STMems_Machine_Learning_Core/tree/master/tools) to be used as example to generate the decision tree in a textual format compatible with Unico GUI.