Skip to content

reichlab/flusight

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

icon


Flusight is a static influenza forecasts visualizer. It takes weekly prediction files in the format used by CDC FluSight Challenge, extracts entities needed for visualization and generates a minified webpage ready for static hosting. The visualizer uses an intermediate file, summarizing the predictions, which can be generated either by the set of scripts in the repository (which uses CDC's prediction format) or by the user. Visualizations are implemented using the d3-foresight npm package.

The live version associated with this github repository at http://reichlab.io/flusight/ displays predictions for CDC FluSight Challenge by Reich Lab at UMass Amherst. To know more about it, go to the about page in the wiki.

Standard - JavaScript Style Guide

DOI Build Status GitHub pull requests

screenshot

Quickstart

For submitting weekly predictions on this repository the recommended method is to create a pull request on the master branch. Fork this project and add your .csv files in ./data/ directory as described in the project’s workflow wiki. The pull requests will be automatically built and once merged to master, will be deployed to the website.

Build it yourself

Flusight works as a static webpage and needs a static hosting service (like github’s). Building the project and bundling data files requires node.js and yarn (or npm).

  • Clone this repository

    git clone [email protected]:reichlab/flusight.git && cd ./flusight

  • Install dependencies

    yarn

  • Put prediction submission files in ./data as described in the wiki

  • Parse prediction files to generate data files in ./src/assets/data/

    yarn run parse

  • Test if the collected data files are fine

    yarn run test

  • Bundle production build

    yarn run build

  • Host it somewhere. Copy over contents from ./dist to your hosting service.

  • Alternatively, a quick deploy to github pages can be done by running

    yarn run deploy

    Remember to point git remote origin to the repository where you want the deployment to take place

In case you want to generate the data files yourself, you can replace the parsing and data collection step with your own but make sure to check for data compliance (yarn run test). Head over to the wiki to know more.

Development

On this repository, commits to master are automatically built and deployed to gh-pages with travis. master contains most recent working version. develop contains the latest commits.

# serve with hot reload at localhost:8080
yarn run dev

# build for production with minification
yarn run build

Go to the development wiki for more details.

Contributing

Contributions are welcome. Use the github issue tracker to submit issues. Submit pull requests here.