File: Subset

package info (click to toggle)
xgobi 19991216-2
  • links: PTS
  • area: non-free
  • in suites: woody
  • size: 3,856 kB
  • ctags: 4,723
  • sloc: ansic: 52,766; makefile: 669; asm: 365; sh: 35
file content (23 lines) | stat: -rw-r--r-- 1,044 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Subset data

Subset the current data in five different ways:

  Consecutive block:  Specify the first and last row of the block
  Random sample without replacement:  Specify the size of the sample
  Every nth row:  Specify the interval and the first row
  Use sticky label ids:  Enter 'Identify' mode and make one or
     more row labels 'sticky.'  This defines the subset of all
     cases having any one of those labels.
  Use row id:  Specify a row label as it would appear when using
    'Identify' to define the subset of all cases with the same row label.

Select one of those five, and set its parameters.  Then click on
"Subset the data" in the bottom row of the popup.  If you want
to re-include all rows, select the "Include all data" button in
the bottom row.

The purpose of this is to enable the use of XGobi on data matrices
that are so large that dynamic and interactive operations begin to
become painfully slow.  By selecting a smaller subset, a user can
work on that subset at a comfortable speed of rotation and
interaction.