Skip to content
Alexander Lex edited this page Jul 29, 2014 · 30 revisions

You can currently provide data to UpSet through a publicly available dataset and a simple description file, which also must be publicly available.

Data Format

UpSet uses a binary encoding for the sets. Here is a simple example, with Sets A, B and C, and three elements in the rows (R1, R2, R3):

Row;A;B;C
R1;1;0;0
R2;0;1;0
R3;0;0;1

You can download this file here.

To make upset understand this data format you have to provide a simple JSON file. The configuration file for the above dataset is as simple as this:

{
	"file": "http://vcg.github.io/upset/data/test/test.csv",
	"name": "Test",
	"header": 0,
	"separator": ";",
	"skip": 0,
	"meta": [
		{ "type": "id", "index": 0, "name": "Name" }
	],
	"sets": [
		{ "format": "binary", "start": 1, "end": 3 }
	]
}

You can download this file here.

The properties of these attributes are the following:

  • file describes the path to the data file. This path typically should be a globally accessible URL, unless you run upset locally, in which case you can use relative paths.
  • name is a custom name that you can give to your dataset, as it will appear in UpSet.
  • header defines the row in the dataset where your column IDs are stored (the sets and the attributes)
  • separator defines which symbols are used to separate the cells in the matrix. Common symbols used are semicolon ;, colon ,, and tab \t
Clone this wiki locally