Sean Owen created SPARK-17768:
---------------------------------
Summary: Small {Sum,Count,Mean}Evaluator problems and
suboptimalities
Key: SPARK-17768
URL: https://issues.apache.org/jira/browse/SPARK-17768
Project: Spark
Issue Type: Bug
Components: Spark Core
Affects Versions: 2.0.1
Reporter: Sean Owen
Assignee: Sean Owen
This tracks a few related issues with
org.apache.spark.partial.(Count,Mean,Sum)Evaluator and their "Grouped"
counterparts:
- GroupedMeanEvaluator and GroupedSumEvaluator are unused, as is the
StudentTCacher support class
- CountEvaluator can return a lower bound < 0, when counts can't be negative
- MeanEvaluator will actually fail on exactly 1 datum (yields t-test with 0 DOF)
- CountEvaluator uses a normal distribution, which may be an inappropriate
approximation (leading to above)
- CountEvaluator, MeanEvaluator have no unit tests to catch these
- Duplication across CountEvaluator, GroupedCountEvaluator
- SumEvaluator might have an issue related to CountEvaluator (or could delegate
to compute CountEvaluator times MeanEvaluator?)
- The stats in each could use a bit of documentation as I had to guess at them
- (Code could use a few cleanups and optimizations too)
--
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
