This Node/Python library builds a model to predict if a particular Pull Request (PR) will be accepted when it is created, by learning information about a Github Project. The aim of this library is to aid Project integrator in managing PRs for a particular project. You can find more information about the model and how in this article.

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

What things you need to install the software and how to install them

You will need the following:

• Python 3.6 or newer
• Node 8 or newer
• MongoDB 3.2 or newer
• Git
• A Github Access Token for using the Github API. This post explains how to get yours.

0. Choose a project to predict. In this document I will use https://github.com/Netflix/pygenie, because it is smaller, but you can use any, like the Node project

1. Clone this repository into your machine:

   git clone https://github.com/sophilabs/pullreq-ml.git

2. (Optional) Install your local copy into a virtual environment. For example using the venv library you can do the following.

   python -m venv venv
   source venv/bin/activate

3. Install dependencies

   cd pullreq-ml # or pullreq-ml-master
   npm install
   pip install -r requirements.txt

4. (Optional) Create a user for your MongoDB instance

   echo "db.createUser({ user: 'github', 'pwd': 'github', roles: ['readWrite'] })" | mongo github

5. Replace the contents of config.js with the actual repo and database authentication. For example

    module.exports = {
        // Local Mongo DB
        MONGO_DB_URL: 'mongodb:https://github:github@localhost:27017/github',
        // Token
        GITHUB_ACCESS_TOKEN: '<your token here>',
        // Repo Information for example for https://github.com/Netflix/pygenie you should put
        REPO_OWNER: 'Netflix',
        REPO_NAME: 'pygenie'
    }

6. Clone the target repo inside the targetrepo folder

   git clone https://github.com/Netflix/pygenie.git targetrepo

7. Start fetching Repo information

   node fetch.js

8. Train and evaluate Pull Request Acceptance for your repository

   python evaluate.py

   You should see an output like the following one

   Report on Test data
             precision    recall  f1-score   support
   
    not merged       0.76      0.22      0.34       264
        merged       0.78      0.98      0.87       753
   
    avg / total       0.78      0.78      0.73      1017
   
   Dumped classifier data to classifier.pkl
   

   This command generates a classifier.pkl binary file which can be used to predict any PR on the target Project.

• Build a file to predict a particular PR against the trained model. A command like:

  > python classify.py https://github.com/nodejs/node/pull/11107
  Will not be merged!

• scikit-learn - Used their algorithms to estimate PR merge predictions.
• MongoDB - Used to store Github downloaded project data.
• Git - Used to compute diffs and analyze PR commit deltas.

Feel free to make a Pull Request if you find a bug or want to implement a feature. We welcome any help.

• Ignacio Avas - Initial work - igui

• Pablo Grill for his insight and knowledge over Machine Learning

pullreq-ml is Copyright (c) 2018 sophilabs, inc. It is free software, and may be redistributed under the terms specified in the license file.

pullreq-ml is maintained and funded by sophilabs, inc. The names and logos for sophilabs are trademarks of sophilabs, inc.