[Python] How to set a fixed random state in RandomizedSearchCV?

I'm using RandomizedSearchCV with RandomForestClassifier in scikit-learn. I want to make sure my results are reproducible across runs. Where should I set the random_state—in the classifier, in RandomizedSearchCV, or both?

Example code:

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import RandomizedSearchCV

clf = RandomForestClassifier()
search = RandomizedSearchCV(clf, param_distributions=params, n_iter=10)


What's the best practice to ensure consistent results?

Continue reading...