SNNS Function Calls

The SNNS function calls control the SNNS kernel. They are available as function calls in batchman. The function calls can be divided into four groups:

• Functions which are setting SNNS parameters :
  -- setInitFunc()
  -- setLearnFunc()
  -- setUpdateFunc()
  -- setPruningFunc()
  -- setSubPattern()
  -- setShuffle()
  -- setSubShuffle()

• Functions which refer to neural nets :
  -- loadNet()
  -- saveNet()
  -- saveResult()
  -- initNet()
  -- trainNet()
  -- testNet()

• Functions which refer to patterns :
  -- loadPattern()
  -- setPattern()
  -- delPattern()

• Special functions :
  -- pruneNet()
  -- execute()
  -- print()
  -- exit()
  -- setseed()

The format of such calls is:

function_name (parameter1, parameter2...)

No parameters, one parameter, or multiple parameters can be placed after the function name. Unspecified values take on a default value. Note, however, that if the third value is to be modified, the first two values have to be provided with the function call as well. The parameters have the same order as in the graphical user interface.

Now a description of the function calls of the first group follows.

• Function Calls To Set SNNS Parameters
  • setInitFunc
  • setLearnFunc
  • setUpdateFunc
  • setPruningFunc
  • setSubPattern
  • setShuffle, setSubShuffle
• Function Calls Related To Networks
• Pattern Function Calls
• Special Functions