[SPARK-13257] [Improvement] Refine naive Bayes example by checking model after loading it #11125
Conversation
|
Please file a JIRA https://cwiki.apache.org/confluence/display/SPARK/Contributing+to+Spark |
|
https://issues.apache.org/jira/browse/SPARK-13257 On Tue, Feb 9, 2016 at 11:28 AM, andrewor14 [email protected]
|
|
Change the title to |
movelikeriver
changed the title
|
|
||
| sc = SparkContext(appName="PythonNaiveBayesExample") | ||
|
|
||
| WORK_DIR = './' |
There was a problem hiding this comment.
This is not necessary. There is a tradeoff between code simplicity and robustness. The main purpose of this example code is to be included in the user guide. See the source code of https://github.com/apache/spark/blob/master/docs/mllib-naive-bayes.md. So simplicity is more important.
There was a problem hiding this comment.
removed. i added this, because it's easy for me to modify the WORK_DIR. agree, simplicity is more important for this sample code.
|
@movelikeriver Thanks for improving the example code! We want to keep the example code simple because they are included in the user guide. So there are some special trade-offs. Also, if you make changes to the Python example code, do you mind updating the Java and Scala examples as well? They should be synced. Thanks! |
|
okay, i'll take a look at java and scala code as well, and create another change soon. |
|
ok to test |
|
Test build #51166 has finished for PR 11125 at commit
|
|
LGTM. Merged into master. Thanks! |
4 participants
Refine naive Bayes example by checking model after loading it