Skip to content

ML data annotations made super easy for teams. Just upload data, add your team and build training/evaluation dataset in hours.

License

Notifications You must be signed in to change notification settings

nkostop/DataTurks

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Features

NER tagging in Documents

Full length document annotations (PDF, Doc, Text etc).

Image Segmentation

    Draw free form polygons and generate image masks.

POS tagging

    A super easy interface to tag for PoS/NER in sentences.

DataTurks

[Note: As on March 2019, please use the branch master_with_dist as the main master branch has some build issues w.r.t dist folder]

Can run as a docker image as well. Here is the docker file specifying all the steps for setting things up:

https://github.com/DataTurks/DataTurks/blob/master/hope/docker/Dockerfile

If you rather have it run as a non-docker service, then see below.

Two main subcomponents:

  1. Hope: Java-mysql based backend.
Build:
Its a maven project, please install maven and then:
# cd hope
# mvn package -DskipTests <-- will build the .jar file.

Run:
The service is based on dropwizard and taken a config file on startup. This config file specifies the MYSQL end-points, 
password and the port to run the service on.

Setup mysql server as in: https://github.com/DataTurks/DataTurks/blob/master/hope/docker/mysqlInit.sql

# java -Djava.net.useSystemProxies=true -server -jar dataturks-1.0-SNAPSHOT.jar server onprem.yml
  1. Bazaar: React based front-end.

Mac Setup :

brew install node@8
brew link node@8
conda create -n bazaar python=2.7 anaconda
conda activate bazaar
xcode-select --install
sudo xcode-select -s /Applications/Xcode.app/Contents/Developer
rm -rf node_modules
npm rebuild node-sass
npm install
npm run dev  

Linux Setup: Install Node Js etc.

sudo apt-get -y install build-essential 
curl -sL https://deb.nodesource.com/setup_8.x | bash 	  
apt-get install --yes nodejs 	  
node -v 	  
npm -v  	  
npm i -g nodemon 	  
nodemon -v	  
apt-get clean 	  

Build:

  cd bazaar
  npm install && npm run build-onprem

Run the service:

  npm run start-onprem

About

ML data annotations made super easy for teams. Just upload data, add your team and build training/evaluation dataset in hours.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 53.4%
  • CSS 23.7%
  • Java 16.1%
  • HTML 5.7%
  • Hack 0.6%
  • PHP 0.2%
  • Other 0.3%