The documentation for RandomizedSearchCV implies that a best_params_ property is available after .fit() is called. This does not appear to be the case.
Here is the documentation in question:
https://github.com/databricks/spark-sklearn/blob/master/python/spark_sklearn/random_search.py#L162
Should I attempt to implement a best_params_ property or indicate in the documentation that best_params_ is not supported (as it is a common API in sklearn-related libraries) in a PR?