home *** CD-ROM | disk | FTP | other *** search
- 1. Processing Examples
- An example run can be made for each processing option in
- this package, using the data files in directory
- \neuron\batch\dat. These processing examples are run as follows.
- a. Go to the subdirectory \neuron\batch\dat
- b. Start up Nuron
- c. Go to the option you are interested in. However, under
- "Multilayer Perceptron Processing", you cannot "Analyze a
- Trained MLP" until you have designed one.
- d. Choose batch processing. The dos editor will edit a parameter
- file, which has responses to programs requests
- e. Exit the Dos editor. The program will now run using responses in
- the parameter file.
- f. Choose the "Examine Program Output" option, if you like.
- g. For further details on the examples, see the help files for
- the option you are interested in.
-
- 2. Designing a MLP Neural Net for Your Own Data
- a. Obtain an ascii data file and put it in the IANS format. File
- Grng is a good example of this format. Figure out how many
- inputs or features your data has, how many classes or possible
- decisions you have.
- b. Decide how many hidden layers (1 or 2) you want and how many
- hidden units (up to 40 per layer) you want. One way to do this
- is to "Estimate Network Structure" under the "Data Format and
- Pre-Processing" option. Create a network structure file by
- editing a copy of Grng.top or during the manual run of the
- backpropagation program or fast training program.
- c. Decide whether to use coded or uncoded format for the output
- layer. Uncoded, in which there is one output per class, is
- recommended.
- d. Decide whether you want to use backpropagation training or fast
- training. Fast training is recommended.
- e. Go to the "Batch Processing" option under "Fast Training" or
- "Backpropagation Training", which are under the "Multilayer
- Perceptron Processing" menu.
- f. Alter the keyboard response file, following the choices you
- have made, and exit the EDIT program. The training program will
- now run.
- g. Choose the "Examine Program Output" option if you like.
-