Closed dberenbaum closed 2 years ago
Using a fractional value for min_split (see https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html) reduces overfitting and scales better to different sample sizes.
min_split
train.min_split=2
train.min_split=0.01
@dberenbaum thanks, I'll regenerate the project with this change.
Using a fractional value for
min_split
(see https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html) reduces overfitting and scales better to different sample sizes.train.min_split=2
train.min_split=0.01