definition of score when using "predict" on trained adaBoostM1

5 Ansichten (letzte 30 Tage)
Neta Gertzovsky
Neta Gertzovsky am 15 Jun. 2016
Beantwortet: Ilya am 15 Jun. 2016
I am using fitensemble to train adaBoost. ('AdaBoostM1'). I can't understand how exactly do they calculate the score in [lbl,score] = predict(ClassTreeEns,X); Is this the value of the classic adaBoost? as in the value of F(x), where F(x)=sigma(alpha_t*f_t(x)), where F is the strong classifier and f_t are the weak classifiers and alpha_t are the weights? I can't find the exact definition of score anywhere in Matlab documentation..

Antworten (1)

Ilya
Ilya am 15 Jun. 2016
Going to MATLAB online doc, typing 'AdaBoost" in the search box and then selecting the 3rd match brings me to this page:

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by