spark-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From BryanCutler <...@git.apache.org>
Subject [GitHub] spark issue #12790: [SPARK-15018][PYSPARK][ML] Fixed bug causing error if Py...
Date Tue, 21 Jun 2016 21:39:30 GMT
Github user BryanCutler commented on the issue:

    https://github.com/apache/spark/pull/12790
  
    Bumping this, there has been some discussion but hopefully we can get a consensus on constructing
a pipeline with no stages (should be an identity trans).  Looking at the current API, it seems
like the following should be supported:
    
    1) `pl = Pipeline(stages=[])`
    2) `pl = Pipeline(stages=None)`
    3) `pl = Pipeline()`
    
    However, only 1) works, while the others will lead to a cryptic error when calling `fit`.
 I think 1) & 3) should be supported, I am indifferent about 2).  Regardless of what's
decided, this code should be cleaned up because the current checks don't do anything, there
is no default value, and `getStages` is missing a return path.
    
    cc @mengxr @jkbradley @MLnick 



---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org


Mime
View raw message