Universal Synaptics Geneprop

	Training Display Analog Graph Patterns Learned Network as Configured Training Metrics All displays programmable Shows training progress

Geneprop A back-propagation Neural Network, Training and Testing Program

Uses for Neural Networks

• Develop Pattern recognition projects
• Perform Data Classification
• Build Decision Support Systems
• Develop mathematical data-driven, curve-fitting functions
• Explore the possibilities of Parallel Distributed Processing
• Learn advanced computer technology

User Friendly screens and menus

• User overhead kept to a minimum
• Setups can be saved and reused for rebuilding
• Training may be stopped and restarted anytime
• Training progress is automatically saved at various milestones
• Suggestions are given for setting up training parameters
• Numerous display and output options
• Paste the results directly into other programs

Main Menu Numerous features and options Totally menu driven Programmable training schedules Enhanced training options Genetic Algorithm selects best patterns to train with Automatic selection of training rates Output product may be written as a function in Basic, C, or Spreadsheet.

Included:

• Documentation
• Complete files from actual projects
• Email support
• Basic source-code for feed-forward program

Requirements:

• PC with 386 or better microprocessor
• Math coprocessor required (486 DX or better)
• 1 Meg disk space
• 640 K conventional memory

To Order

E-Mail: sales@usynaptics.com



Universal Synaptics Corporation
1801 West 21st Street
Ogden, Utah 84401 USA
Tel. (801) 731-8508
Fax (801) 731-8509