Generate Neural Network Input Files

The first part of this process is the same as Calculating a Response Theme for the weights of evidence and logistic regression methods.

  1. Select 'Generate Neural Network Input Files...' from the SDM menu. This displays the 'Input to Neural Network - Themes' dialog.
  2. Select the evidential themes to include in the analysis.
  3. Click 'Specify Fields'. This opens the 'Inputs to Neural Network - Fields' dialog.
  4. For each evidential theme, specify the field containing the class values to analyse, the integer that defines areas of missing data and the theme data type, either free or ordered.
  5. Click OK.
  6. Click 'Generate Input Files...'.
  7. When prompted, specify the following file names and locations:

filename and location of the...

Description

Default Name
unique conditions theme grid Jump to definition. sdmuc#
training file
  • only generated if the RBFLN option is selected
  • text file containing information from unique conditions in which training points are located
  • 1 row = 1 unique condition
  • in DataXplore, 1 row = 1 training vector
  • each unique condition is written once, even though it may contain more than one training point
  • if a training point indicating presence of an object and another indicating absence occur in the same unique condition, the training point indicating presence takes priority
 train#.dta
data file
  • text file containing complete unique conditions
  • 1 row = 1 unique condition
  • in DataXplore, 1 row = 1 feature vector
 class#.dta

After the files have been created, the unique conditions grid theme will be added to the active view with a default name of 'Neural Network #'.

Sample input files.

Unique conditions table in ArcView
Neural network input Files
A. Unique conditions table written to neural network input file, delimited text format
A. Unique conditions table written to neural network input file, delimited text format
Description of file header:

Line 1 (5): Number of evidential themes.
Also called components or features, in the context of neural networks and DataXplore.

Line 2 (151): Number of training points.
In the context of neural networks, "the number of centres (for clustering) or the number of radial basis functions (each with a centre) when our neural networks are used. But the number is not fixed, and may be changed by the program or by the user in either the unsupervised (fuzzy) clustering or in the supervised training of radial basis function neural networks.

Line 3 (1): Number of output components, or values to be mapped as response themes.
This is always set to one in the Arc-SDM application.

Line 4 (A: 3462) (B: 151): Number of unique conditions.
In the context of neural networks, the number of target vectors.

Description of data columns:

Column 1: Unique condition number.

Column 2: Number of training points in unique condition (not currently used).

Column 3: Area of unique condition (not currently used).

Columns 4 - 8: Unique condition values or contents of target vectors.
Values are transformed from the ArcView unique conditions table in two ways:

  1. the range of an evidential theme's values is normalized between 0 and 1
  2. the missing data integer is replaced with an area weighted mean

Column 9: The output component

 

Next Section Contents Home