bugbug
bugbug copied to clipboard
Platform for Machine Learning projects on Software Engineering
It can be an option of the model.
A bit like the test selection model.
https://xgboost.readthedocs.io/en/latest/tutorials/categorical.html
If the list of all S2 bugs is too large, we can go with new S2s only, or new S2s and S2s with no activity for X weeks (new S2s...
Currently the quality report notification carries a link to the bugbug UI. The there shown stats are interesting but quite different from those of the email. This is - confusing?...
Try restricting the training set for the regression model to defects only (regression defects vs non-regression defects, instead of regressions vs non-regressions defects+non-defects).
The spambug model is only applied to bugs filed by people without "editbugs" permissions, so it makes sense to only evaluate it on these kinds of bugs and not all...
We apply the model to bugs from Thunderbird too for example, but we only train on Firefox, Toolkit, etc. bugs. This means Thunderbird bugs are more likely to be wrongly...
These components might no longer be used if there are no recent bugs in them.