Hello, can you tell me which method of joint learning the random forest classifier of the OTB complement uses, boosting or bagging?
I did not find such information in the documentation.
Hi @ViGeo
What RF classifier do you use ? sharkrf or rf, nevertheless i would say boosting by looking at the code but I have to verify with the classifier you are using (rf is using openCV, sharkrf is using Shark).
By the way, In OTB you can explicitly call the boosting method using classifier.boost, which calls opencv boosting method, the doc part in the cookbook is here
Best regards
Hello,
I’m using Random forests classifier.
I ask so I can quote correctly in the article.
Hello @ViGeo
I looked further in the code and asked a contributor who worked on that, he confirmed that the random forest in OTB uses the bagging method. This is why there is a Boosting classifier which specifically does the boosting method.
Best regards