Correct me if I am wrong, oob_score is the score of random forest regressor on that validation set created by the unions of all the rows left-out after each tree training set.
Yes, I guess. When you take a sample with replacement, some rows will be left out during the training phase of each tree. Therefore we can think of getting those rows and compute the score with those. I do not know if it is appropriate to define it a ‘validation set’, though!
Anyone knows how to change the
oob_score_ in sklearn’s randomforest from the default accuracy to some other metric, e.g F1, precision, recall, ROC etc? there is absolutely nothing on this when one google search…is it too dumb or no one ever needed to do so?