You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Dec 9, 2022. It is now read-only.
Repository navigation
This repository was archived by the owner on Dec 9, 2022. It is now read-only.
Want to see a dry run of predictions on existing issues before feeling comfortable installing this.
The bot should say "I am not sure" instead of staying completely silent when it is not sure about a label so that the user doesn't end up wondering if the bot is working or not.
Can the bot respond in situations where the title change or more comments are added?
Redirect fromt http to https incase someone types in the wrong address
Have sandbox repo for users to test things out on
Added ability to alias the name of the labels for example "#bug" instead of "bug"
cc: @inc0@jlewi@rand@jessfraz please provide feedback and add more ideas here please, and invite others to make suggestions on features as well!
Issue-Label Bot is automatically applying the label feature_request to this issue, with a confidence of 0.78. Please mark this comment with 👍 or 👎 to give our bot feedback!
@hamelsmu would it be possible to drive the dry run off of issues on this repo? Give potential users a prompt to open an issue here, and see how the results/flow work?
@rand that wouldn't address my concern which is that i have no idea what this bot is going to do on a big repo (e.g. electron/electron). What tags is it going to add? How often will it be correct? What's the false positive / false negative rate?
Those questions are not answered by seeing how it works on a different repo.
@nornagon ah, thank you for the clarification of your concerns. To confirm: you're looking for a preview + summary of what this would do on a specific repo, rather than just a test environment?
Getting the false positive/negative rates beforehand will be challenging, as that data is accrued via 👍 👎 feedback on the bot's interactions.
As for dry run, we could also provide webpage where you could input your issue link and it will show prediction. People would be able to just try it out with their issues
Could you perform a con-sine similarity or similar to identify and aggregate issues open/new? Then with some confidence issues could be closed as dupe?
@DevAvanCodes I opened a PR on your repo: https://github.com/DevAvanCodes/YourSkinGUI/pull/1 as an example of how to alias one of the labels, Labeling "feature_request" as "enhancement" -> we can take the conversation there perhaps if there are any questions?
@DevAvanCodes I opened a PR on your repo: https://github.com/DevAvanCodes/YourSkinGUI/pull/1 as an example of how to alias one of the labels, Labeling "feature_request" as "enhancement" -> we can take the conversation there perhaps if there are any questions?
@hamelsmu Do you think it makes sense to, by default, rename "feature_request" to "feature request"? GitHub issue labels support spaces in the name, like the default "good first issue". I understand that you can alias the labels, but I'm curious about what you think about removing the underscore by default.
Activity
issue-label-bot commentedon Apr 11, 2019
Issue-Label Bot is automatically applying the label
feature_requestto this issue, with a confidence of 0.78. Please mark this comment with 👍 or 👎 to give our bot feedback!Links: dashboard, app homepage and code for this bot.
rand commentedon Apr 11, 2019
@hamelsmu would it be possible to drive the dry run off of issues on this repo? Give potential users a prompt to open an issue here, and see how the results/flow work?
The rest of the feedback seems spot on.
hamelsmu commentedon Apr 11, 2019
@rand thats a great suggestion to have a sandbox repo to test things out on I like it
nornagon commentedon Apr 11, 2019
@rand that wouldn't address my concern which is that i have no idea what this bot is going to do on a big repo (e.g. electron/electron). What tags is it going to add? How often will it be correct? What's the false positive / false negative rate?
Those questions are not answered by seeing how it works on a different repo.
rand commentedon Apr 11, 2019
@nornagon ah, thank you for the clarification of your concerns. To confirm: you're looking for a preview + summary of what this would do on a specific repo, rather than just a test environment?
Getting the false positive/negative rates beforehand will be challenging, as that data is accrued via 👍 👎 feedback on the bot's interactions.
nornagon commentedon Apr 11, 2019
Yeah, I'm not looking for aggregate statistics, I just want to see how it behaves on a few sample issues and see if it seems sane or not.
hamelsmu commentedon Apr 11, 2019
@nornagon I'll think about the best way to handle this and then tag you again. Thank you for the feedback!
hamelsmu commentedon Apr 11, 2019
@nornagon I copied pasted some issues for you in https://github.com/hamelsmu/Issue-Label-Bot-Examples/issues, feel free to give it a spin if you want and open more issues!
I'll add some kind of feature that allows you do this but maybe this will get us bootstrapped in the meantime? Let me know your thoughts!
inc0 commentedon Apr 15, 2019
As for dry run, we could also provide webpage where you could input your issue link and it will show prediction. People would be able to just try it out with their issues
nornagon commentedon Apr 15, 2019
That sounds useful.
kwartler commentedon Apr 18, 2019
Could you perform a con-sine similarity or similar to identify and aggregate issues open/new? Then with some confidence issues could be closed as dupe?
hamelsmu commentedon Apr 18, 2019
@kwartler that is an excellent idea for another app! We have some tech that we could bootstrap for this purpose! (see this)
hamelsmu commentedon May 4, 2019
@DevAvanCodes can you point me to one of your repos and what you are trying to do so I can take a look?
hamelsmu commentedon May 5, 2019
@DevAvanCodes I opened a PR on your repo: https://github.com/DevAvanCodes/YourSkinGUI/pull/1 as an example of how to alias one of the labels, Labeling "feature_request" as "enhancement" -> we can take the conversation there perhaps if there are any questions?
AnandChowdhary commentedon Jun 12, 2019
@hamelsmu Do you think it makes sense to, by default, rename "feature_request" to "feature request"? GitHub issue labels support spaces in the name, like the default "good first issue". I understand that you can alias the labels, but I'm curious about what you think about removing the underscore by default.