A file format for double indexed data

**TABLE** is a data directory which
contains examples of TABLE files, which are a simple
homegrown storage format for doubly indexed data.

A TABLE file has a simple structure. There are three kinds of records:

- COMMENT LINES begin with the character '#' and are ignored;
- BLANK LINES are ignored, but could be used by a particular program to indicate some intending grouping or separation of records.
- COORDINATE LINES each contain one set of point coordinates;

The computer code and data files described and made available on this web page are distributed under the GNU LGPL license.

Example: 3 points in 2D

```
# Corners of a triangle
#
0.0 0.0
1.0 0.0
0.0 1.0
```

Example: 4 points in 3D

```
# Corners of a tetrahedron
#
0.0 0.0 0.0
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
```

Example: 3 points in 7D

```
# Three points in 7 dimensions.
#
0.7 0.2 0.2 0.3 0.5 0.6 0.3
1.9 1.3 1.2 2.2 2.2 2.2 9.1
0.9 0.7 1.1 0.4 0.8 2.7 0.0
```

Two special cases of this format are

- ASCII
- No compression
- 1 "image"

- The CVT dataset collection, tables of pointsets in M dimensions, generated by the centroidal Voronoi tessellation technique;
- The CVT MOD dataset collection, tables of pointsets in M dimensions, generated by the centroidal Voronoi tessellation technique with a special modularized geometry that eliminates boundary effects;
- diamond_02_00009.txt, 9 points in 2D, whose Voronoi diagram includes a diamond;
- The FAURE dataset collection, tables of pointsets in M dimensions, generated by the Faure quasirandom sequence;
- The GRID dataset collection, tables of pointsets in M dimensions, generated by choosing (some) points on a regular grid;
- The HALTON dataset collection, tables of pointsets in M dimensions, generated by the Halton quasirandom sequence;
- The HAMMERSLEY dataset collection, tables of pointsets in M dimensions, generated by the Hammersley quasirandom sequence;
- The HEX GRID dataset collection, tables of pointsets in M dimensions, generated by constructing a hexagonal grid in a rectangle;
- The Improved Hyper Square dataset collection, tables of pointsets in M dimensions which are Latin hypercubes, with a shuffling scheme applied to improve the distribution;
- The LATIN CENTER dataset collection, tables of pointsets in M dimensions which are Latin hypercubes, with each point chosen in the center of its subsquare;
- The LATIN EDGE dataset collection, tables of pointsets in M dimensions which are Latin hypercubes, with the point chosens as the corners of the subsquares;
- The LATIN RANDOM dataset collection, tables of pointsets in M dimensions which are Latin hypercubes, with each point chosen randomly from a subsquare;
- The LCVT dataset collection, tables of pointsets in M dimensions, generated by the "latinized" CVT technique;
- The LCVT MOD dataset collection, tables of pointsets in M dimensions, generated by the "latinized" CVT technique with a MOD operation employed to eliminate boundary effects;
- The LHS dataset collection, tables of pointsets in M dimensions, generated by the Latin Hyper Square system;
- The NIEDERREITER2 dataset collection, tables of pointsets in M dimensions, generated by the Niederreiter quasirandom sequence with base 2;
- The orthogonal array collection, tables of integer matrices, which are a generalization of Latin squares;
- The Romero dataset collection, tables of pointsets in M dimensions, generated by the "latinized" CVT technique;
- The SAMPLE2D dataset collection, tables of pointsets in M dimensions, generated by a variety of techniques;
- The SOBOL dataset collection, tables of pointsets in M dimensions, generated by the Sobol quasirandom sequence;
- the SPAETH dataset, a set of files of data for demonstrating clustering techniques.
- the SPAETH2 dataset, a set of files of data for demonstrating clustering techniques.
- tetrahedron_03_00004.txt, 4 points in 3D, which form a tetrahedron;
- The UNIFORM dataset collection, tables of pointsets in M dimensions, generated by a a simple uniform pseudorandom number generator;

You can go up one level to the DATA page.