A truth inference tool in crowdsourcing
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
CrowdInfer
main
truth_inference_crowd
upload
.gitignore
Main.iml
README.md
main_method.sql
manage.py

README.md

CrowdTI

Dependency

celery

Follow instructions at here

Briefly you can use brew to install rabbitmq needed by celery.

brew install rabbitmq

After installing, you should alter environment variable PATH as below:

export PATH=$PATH:/usr/local/sbin

Then install celery using pip

pip install celery

If you have problems with celery and django, you can refer to this

gviz_api

pip install -U https://github.com/google/google-visualization-python/zipball/master

mono

Please refer here to install mono, in order to compile and run .NET application.

MATLAB

Please install MATLAB and make sure you can open it in terminal using 'matlab'.

jquery.timers

Have already contained in project. More info here.

Database configuration

The project used django default database, you can use your own as long as you configure it in CrowdInfer/settings.py, you can refer to here for instructions.

Tested Environment

####OS X 10.11

  • Python 2.7.10
  • Django 1.10.1
  • rabbitmq 3.6.1
  • celery 4.0.2
  • mono 4.2.3
  • MATLAB R2014b

Make it Run

Start rabbitmq

Start celery

celery -A CrowdInfer worker -l info

Migrate Django Tables

python manage.py migrate

Then you should load method info from 'main_method.sql'

Run Django

python manage.py runserver

Troubleshooting

If you're encountered with error like

DisabledBackend' object has no attribute '_get_task_meta_for

Make sure you have prepared CrowdInfer/init.py as instructions.