How to get probabilities from Boosting algorithms?


#1

Hi,
I am currently using the R logistic regression to provide me with coefficients later plugged
( this done in a vba macro) into a logistic equation which provide probabilities for race horses.
In order to get better results,I would like to do the same thing but using a boost method, which would also provide either coefficients or probabilities.
Has anybody any advice on how to proceed?
sandoz


#2

Hi @sandoz,

In python you can simply do

model.predict_proba(X, y)

using scikit-learn to get probabilities from classification models, this includes Adaboost and Gradient Boosting algorithms


#3

And in “R”:

predict(your_model, newdata=data, class = “prob”)

Thanks,
Carlos Ortega


#4

thank you, this will help me !