There are still methods, which are using toPandas(), which is not the best way when running on Databricks. Given there is no reason not to use PySpark all of these expectations should be changed.