From Chorus

How to use this template

This template should be used in all pages describing datasets.

What this template contains

url

Please provide a wiki-formatted link to the source of the dataset, e.g.

 url = http://www.example.com/some/path

 url = [http://www.example.com/some/improtant/file/hidden/behind/a/long/path.pdf http://www.example.com/.../path.pdf]

domain

The domain this dataset comes from, e.g. Medical, News, Picture. The corresponding Category will be automatically added.

 domain = News

media

Possible values: Image, Audio, Video or Text

The page will be automatically added to the corresponding Category.

 media = Image

size

The size of the dataset (in MB/GB)

 size = 150 GB

instances

Number of instances (data items) in the dataset

 instances = 153000 songs

format

The file format (file extension) in upper case. The corresponding Category will be automatically added.

 format = TXT

created

The creation date of the dataset in the format YYYY-MM-DD

 created = 2010-12-24

task

The task this dataset is typically used: Retrieval, Classification or Recognition. The corresponding Category will be automatically added.

 task = Classification

copyright

What are the copyright limitations of this dataset