zhengruifeng created SPARK-19208:
------------------------------------
Summary: MaxAbsScaler and MinMaxScaler are very inefficient
Key: SPARK-19208
URL: https://issues.apache.org/jira/browse/SPARK-19208
Project: Spark
Issue Type: Improvement
Components: ML
Reporter: zhengruifeng
Now, {{MaxAbsScaler}} and {{MinMaxScaler}} are using
{{MultivariateOnlineSummarizer}} to compute the min/max.
However {{MultivariateOnlineSummarizer}} will also compute extra unused
statistics. It slows down the task, moreover it is more prone to cause OOM.
For example:
env : --driver-memory 4G --executor-memory 1G --num-executors 4
data: 748401 instances, and 3,000,000 features
{{MaxAbsScaler.fit}} fail because OOM
{{MultivariateOnlineSummarizer}} maintain 8 arrays
For {{MaxAbsScaler}}, only one array is needed (max of abs value)
For {{MinMaxScaler}}, only 3 arrays are needed (max, min, nnz)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
