Skip to content

A Python Machine Learning Pipeline to predict software defects in java projects

Notifications You must be signed in to change notification settings

ML-Java-Code-Analysis/ML-Pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ML-Pipeline

This is the Machine-Learning-Pipeline for the bachelor thesis project "Recognizing defects in Java code automatically".

Setup

Python 3.5 with scikit-learn and SQLAlchemy is required. These libraries can easily be installed with Anaconda.

For Windows:

  1. install Anaconda
  1. Setup environment
  • Enter into cmd (Windows):

  • conda create --name ml scikit-learn sqlalchemy pymysql matplotlib

    activate ml

    pip install terminaltables

  1. Configure PyCharm:
  • Open settings (CTRL + ALT + S)
  • Project: ML-Pipeline -> Project Interpreter
  • Click the cog wheel next to the interpreter listbox -> Add Local
  • Choose the Python.exe of your new environment
    • E.g. C:\Anaconda3\envs\ml\python.exe
  • Now all packages like SQLAlchemy, scikit-learn, numpy etc. should be listed.
  • Be patient, PyCharm needs some time to rebuild its indexes -> In the lower right corner PyCharm tells you which processes are running.

ML-Pipeline

Die Machine-Learning-Pipeline für das Bachelorarbeitsprojekt "Fehler in Java Code automatisch erkennen".

Setup

Es wird Python 3.5 mit Scikit-learn und SQLAlchemy benötigt. Am einfachsten geht die Installation mit Anaconda:

  1. Anaconda installieren
  1. Environment aufsetzen:
  • In der Kommandozeile (Windows):

  • conda create --name ml scikit-learn sqlalchemy pymysql matplotlib

    activate ml

    pip install terminaltables

  1. PyCharm konfigurieren:
  • Settings öffnen (CTRL + ALT + S)
  • Project: ML-Pipeline -> Project Interpreter
  • Auf Zahnrad neben Interpreter Listbox klicken -> Add Local
  • Python.exe des neuen Environments auswählen
    • Z.B. C:\Anaconda3\envs\ml\python.exe
  • Nun sollten alle Packages wie SQLAlchemy, scikit-learn, numpy etc. aufgelistet sein.
  • Dann geduldig sein, PyCharm hat eine Weile weil es die Indizes und so neu bilden muss -> Rechts unten in PyCharm steht, was für Prozesse am laufen sind.

About

A Python Machine Learning Pipeline to predict software defects in java projects

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages