3.3 Importing Data
Of the data formats that Data Explorer can import, the General Array format is
likely to be the most useful to a majority of users.
Other formats are discussed
in detail in Appendix B. "Importing Data: File
Formats"
in IBM Visualization Data Explorer User's Guide.
The General Array format uses a simple header file to describe data
characteristics such as grid dimensions, data type, and layout.
The two examples here use the Data Explorer Data Prompter to create such header
files
for two simple data files.
The Data Prompter is designed specifically for importing data in General
Array format.
(For details of format and the Data Prompter, see Chapter 5. "Importing Data".)
Example 1
This example illustrates the importation of a simple data set consisting
of a single variable on a 5 × 5 × 5 grid.
- First enter:
dx -prompter
You can also access the Data Prompter by choosing Import
Data from the Data Explorer Startup window.
When the initial dialog box appears (see Figure
15),
choose Grid or Scattered File, and then type the path name
/usr/lpp/dx/samples/tutorial/external.data.
into the Data file name field at the top of the
dialog.
- Press the Describe Data button to bring up
a window which allows you to describe the data.
- You can now view the file by clicking on the ellipsis button
(...) to the right of the file-name field and
selecting Browser from the
pull-down menu.
The File Browser window appears.
(If necessary, move this window so that you can view it and the Data Prompter
window at the same time.)
It is also possible to browse the data file from the initial prompter
window once the file name has been entered.
- The first three lines of the file may look like data, but they are
header information, so click on the Header
toggle button and then on # of bytes
to the right (now activated).
- Select # of lines and enter the value "3"
in the associated field.
- The first line of the header gives the grid dimensions as
5 5 5, so enter these numbers in the first three
Grid size fields.
Note what happens in the Grid positions fields as
each number is entered.
(You can use the tab key or the mouse to move the cursor from one field
to the next.)
- For Data format,
ensure that ASCII (text) is selected.
- Next you need to specify whether the data is in row or column
majority order.
In this particular file, the data is in row majority, so select
that button.
- Position the cursor in the first origin, delta
field, hold down the left mouse button, and drag the cursor over
the numbers there to highlight the field before releasing the
mouse button.
The default values and the highlight bar will disappear as soon as you
start typing.
- The origin of the grid is [1.0, 3.0, 2.0] (second line of
the header) and the corresponding deltas are .5, .3, and .8,
respectively (third line).
So enter the values:
1, .5
3, .3
2, .8
in the three origin, delta fields.
- On the right side of the prompter window, you can change the name
of the data variable (field0 by default) and
specify the data type and structure.
For this example, the data are floating-point scalar, so you do not
need to change the settings.
- You can now save the header file you have defined.
Select Save As... from the
File
pull-down menu.
Save the file under any name you choose.
Note: The data can now be imported by specifying this file name to an
Import module.
However, if the extension is not "general," you must specify
"general" in the "format" parameter field of the
Import configuration dialog box
(see "Controlling Inputs: Configuration Dialog
Boxes").
- In the initial Data Prompter window, note that the Test
Import and Visualize Data buttons are
now
enabled. First choose Test Import. A window appears,
displaying a description of the imported data. Now choose Test
Import. A general purpose visualization program will be run
on this data set. To view the program, choose Open Visual Program Editor
from the Windows menu of the Image window. You can also experiment with
changing the interactor settings in the control panel.
Example 2
In this example, you will use the initial dialog box to customize the
Data Prompter before importing a data file that contains scattered data
values for two variables.
The organization of the file is:
x, y, data1, data2
x, y, data1, data2
. . .
where x, y define the positions (or locations) of the data.
(See also "For Future Reference" later in this example.)
- First, invoke the Data Prompter (as in Example 1,
Step 1).
When the initial dialog box appears, click on the
Grid or Scattered File button for
scattered data. Then
click on the Grid Type button
for scattered data (farthest right).
- Use a stepper button to set Number of variables
to "2."
- The data positions are specified in the data file itself, so
activate the Positions in data file toggle.
- The data positions are 2-dimensional (x,y), so use a stepper
button to set Dimension to "2."
- Verify that the Single time step toggle is
activated.
- Set Data Organization to
Columnar.
- Click on
Describe Data.
The simplified prompter window appears.
- Position the mouse cursor in the Data file field
at the top left and type in the path name
/usr/lpp/dx/samples/tutorial/spreadsheet.data.
You can view the file by choosing Browser (as in
Example 1).
- There is no header in this file.
Set # of Points to "49."
- To save the header file you have defined, select Save
As... in the File pull-down menu.
Save the file under any name you choose.
- The Test Import and Visualize
Data buttons in the initial Data Prompter window are now
enabled. Choose Visualize Data to see a
visualization of this scattered data.
Many other examples of using the General Array format can be
found in 5.1 , "General Array Importer".
Data Prompter options are described in 5.4 ,
"Data Prompter".
| For Future
Reference |
|---|
|
It is important to note that the top-to-bottom order of items in the
Field list (right side of window) is the
same as the left-to-right order of items in the
data file itself:
locations x, y
Field0 data1
Field1 data2
... ...
("locations" is a General Array reserved word used to indicate
when numbers in a data file are to be interpreted as
"positions").
The order of the field list can be changed (with the Move
field stepper buttons) to agree with the order
in the data file.
If other information (e.g., descriptive text) is interspersed among the
data values, you must use the layout options available in the full
Data Prompter (see Layout
and "layout").
|
