Skip to content

PeerProducedResearch/tweetannotator

Repository files navigation

TweetAnnotator

A simple Django website that allows people to annotate which crawled tweets that supposedly are about current COVID-19 symptoms are actually describing them. The whole annotation pipeline is supposed to work out in the open: Participants can

Setup

Run the following (assuming you already have pipenv and heroku installed)

pipenv install
pipenv shell
heroku local:run python manage.py migrate
heroku local:run python manage.py import_tweets
heroku local

Your website should be up at 127.0.0.1:5000.

i18n

The site is already setup to work multi-lingual. If you change any of the template texts you can get create updated translation files with

heroku local:run python manage.py makemessages

If you use Transifex for doing the translation do

tx push -s

to push the latest source language (english). Once the translations are done you can get them and compile for usage via:

tx pull
heroku local:run python manage.py compilemessages

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published