Batchman

Since training a neural network may require several hours of CPU time, it is advisable to perform this task as a batch job during low usage times. SNNS offers the program batchman for this purpose. It is basically an additional interface to the kernel that allows easy background execution.

• Introduction
  • Styling Conventions
  • Calling the Batch Interpreter
• Description of the Batch Language
  • Structure of a Batch Program
  • Data Types and Variables
  • Variables
  • System Variables
  • Operators and Expressions
  • The Print Function
  • Control Structures
    • The If Instruction
    • The For Instruction
    • While and Repeat Instructions
• SNNS Function Calls
  • Function Calls To Set SNNS Parameters
    • setInitFunc
    • setLearnFunc
    • setUpdateFunc
    • setPruningFunc
    • setSubPattern
    • setShuffle, setSubShuffle
  • Function Calls Related To Networks
  • Pattern Function Calls
  • Special Functions
• Batchman Example Programs
  • Example 1
  • Example 2
  • Example 3
• Snnsbat -- The predessor
  • The Snnsbat Environment
  • Using Snnsbat
  • Calling Snnsbat