~99% of GitHub issues for Keras are user errors, making it quite difficult to surface any potential bug.
I wish there was a community workflow for issue escalation.
-
-
wouldn't labels work for marking actual bugs, so one can filter them
-
exactly. Look at https://github.com/matplotlib/matplotlib/issues …. They handle the issue load quite well.
End of conversation
New conversation -
-
-
@Smerity how about an AI to triage issues? ;) -
:) I have a simple model to group issues by 3 labels, tasks, chores and bugs. but Human Language is hard to categ.
End of conversation
New conversation -
-
-
a community norm that one should create a reproducible example when reporting a bug is a start http://bl.ocks.org/mbostock/370b737ce71bed0749e103b01ce1bfaf …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
a good bunch of them have answers in them or "my bad nvm". Is there a way to delegate issues cleaning on github ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Train NNet to classifie real issues from "fake issues"?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
1. be more aggressive with closing "Help me with" issues 2. add "ask for help in gitter" in all caps to issue template
-
also 3. Use github bots for setting triage/bug/feature request/etc labels. Workflow for issue "triage-> close/actual label ->fix"
End of conversation
New conversation -
-
-
Recommend posting to StackOverflow before raising an issue, perhaps? Requires MVCE and takes burden off contributors.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.