SparkML pipeline training slow for small size data but constraint propagation key cannot be disabled on pyspark?

后端 未结 0 1656
一个人的身影
一个人的身影 2021-02-02 15:44

I am trying to speed up the Spark ML pipeline on databricks.

Currently, it cost 35 mins to train a SparkML pipeline for a dataframe with 2000 rows and 3 columns.

B

相关标签:
回答
  • 消灭零回复
提交回复
热议问题