You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Ryosuke Horiuchi (Jira)" <ji...@apache.org> on 2022/06/16 07:46:00 UTC
[jira] [Updated] (SPARK-39446) Add relevance score for nDCG evaluation in MLLIB
[ https://issues.apache.org/jira/browse/SPARK-39446?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Ryosuke Horiuchi updated SPARK-39446:
-------------------------------------
Description:
The current implementation of ndcgAt function treats relevance score as binary (as written [Document|[https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt]|https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt]).
However, it is better to extend this to accept a user-defined relevance score to calculate nDCG more flexibly.
was:
The current implementation of ndcgAt function treats relevance score as binary (as written [Document|[https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt]|https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt].]
However, it is better to extend this to accept a user-defined relevance score to calculate nDCG more flexibly.
> Add relevance score for nDCG evaluation in MLLIB
> ------------------------------------------------
>
> Key: SPARK-39446
> URL: https://issues.apache.org/jira/browse/SPARK-39446
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Affects Versions: 3.2.1
> Reporter: Ryosuke Horiuchi
> Priority: Major
>
> The current implementation of ndcgAt function treats relevance score as binary (as written [Document|[https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt]|https://spark.apache.org/docs/latest/api/python/reference/api/pyspark.mllib.evaluation.RankingMetrics.html#pyspark.mllib.evaluation.RankingMetrics.ndcgAt]).
> However, it is better to extend this to accept a user-defined relevance score to calculate nDCG more flexibly.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org