mozilla
/
bugbug
зеркало из https://github.com/mozilla/bugbug.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
Platform for Machine Learning projects on Software Engineering
77 коммитов 20 Ветки 541 Тег 406 MiB
Python 90.2%
JavaScript 7.2%
HTML 1.7%
CSS 0.7%
Shell 0.2%
Перейти к файлу
Marco Castelluccio 1f5fa66957 Add script to run the training 2018-11-20 16:47:24 +01:00
bugbug Add option to store trained model 2018-11-20 16:47:15 +01:00
labels Move labels into a subdirectory 2018-11-20 00:10:56 +01:00
tests Add tests for the 'labels' module 2018-11-20 11:53:41 +01:00
.gitignore Ignore extracted bugs.json and commits.json files 2018-11-12 12:56:09 +01:00
.isort.cfg Enable several flake8 checkers 2018-09-21 16:45:04 +02:00
.travis.yml Automatically download databases when they don't exist 2018-11-20 14:14:55 +01:00
LICENSE First commit 2018-03-11 20:12:35 +00:00
MANIFEST.in Add setup.py and test building bugbug package (#1) 2018-11-19 22:53:17 +01:00
README.md Create 'data' directory before downloading databases 2018-11-20 11:53:47 +01:00
VERSION Add setup.py and test building bugbug package (#1) 2018-11-19 22:53:17 +01:00
handwritten_rules_run.py Rename get_labels to get_bugbug_labels since labels can now include multiple kinds of labels 2018-11-20 01:20:38 +01:00
requirements.txt Update dependencies (#2) 2018-11-20 16:19:10 +01:00
run.py Add script to run the training 2018-11-20 16:47:24 +01:00
setup.cfg Enable several flake8 checkers 2018-09-21 16:45:04 +02:00
setup.py Add setup.py and test building bugbug package (#1) 2018-11-19 22:53:17 +01:00
test-requirements.txt Update dependencies (#2) 2018-11-20 16:19:10 +01:00

README.md

bugbug - Classify Bugzilla bugs between actual bugs and bugs that aren't bugs

Bugs on Bugzilla aren't always bugs. Sometimes they are feature requests, refactorings, and so on. The aim of this project is to distinguish between bugs that are actually bugs and bugs that aren't.

The dataset currently contains 2110 bugs, the accuracy of the current classifier is ~93% (precision ~95%, recall ~94%).

Setup

  1. Run pip install -r requirements.txt and pip install -r test-requirements.txt
  2. Run mkdir data.
  3. Run wget https://www.dropbox.com/s/mz3afgncx0siijc/commits.json.xz?dl=0 -O data/commits.json.xz
  4. Run wget https://www.dropbox.com/s/xm6wzac9jl81irz/bugs.json.xz?dl=0 -O data/bugs.json.xz

If you update the bugs database, run xz -v9 -k data/bugs.json. If you update the commits database, run xz -v9 -k data/commits.json.