Save a random tree in python and load it in C++

edit

sure that should be possible. on the python side it is:

rf = cv2.ml.RTrees_create()
# ... train it
rf.save("rf.yml")
# also see:
# help(cv2.ml.RTrees_create())

on the c++ side it is:

// note, that loading a model creates a new object (in both py and c++):
Ptr<ml::RTrees> rf = ml::StatModel::load<ml::RTrees>("rf.yml");
float p = rf->predict(something);