Help Topics
Concepts Package
Class
The Import Bean panel provides these options:
- Definition File Name
- Enter the name of a definition
file. Use the Browse button to select a
definition file. This file can be used to generate a Filter bean
The data itself must be a file by the same name with a .dat
extension.
- Delimiters
- Select one of the following delimiters:
Spaces use spaces between fields.
Commas use commas between fields.
- Record cache
- Enter the number of records to be read from the file and
stored in memory. To read the entire file and store in
memory, enter a 0 value.
- Randomize data
- Check this box if the file data should be presented in a
random sequence. For some beans such as BackPropagation,
this can be beneficial; other beans such as Temporal
Difference Learning the record sequence is critical.
The Import Bean panel is used to specify and operate on a file
from which data is to be read. It is often used before a filter bean to translate data
for neural network input.
Steps in using the panel include:
- Enter the name of the import file from which to read data.
- Select Spaces or Commas as present in the
data file.
- Check Randomize data if beneficial to the
receiving bean.
- Press the Open file button.
- Press the Generate Filter button to create a
translate file with the same name as the definition file.