akshayrai commented on a change in pull request #4423: [TE] add ml auto config
into YAML template
URL: https://github.com/apache/incubator-pinot/pull/4423#discussion_r303559300
##########
File path: thirdeye/thirdeye-frontend/app/utils/constants.js
##########
@@ -27,6 +27,7 @@ rules: # Can configure multiple
rules with "OR" relat
configuration:
bucketPeriod: P1D # Use PT1H for hourly and PT5M for minute
level (ingraph metrics) data.
pValueThreshold: 0.05 # Higher value means more sensitive to small
changes.
+ mlConfig: true # The machine learning auto config to select
and maintain the configuration with the best performance.
Review comment:
Putting this in the template will not guarantee that very new alert will
have this setting. Users will also copy paste configs from wiki and other
alerts. We can monitor the performance by cloning some of our existing alerts
right?
I would still prefer testing it offline and setting the default to true
rather than an optional field in the template. Otherwise, I think it we can
document it in the wiki rather than the template and keep the configuration
simple.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
