Description

This Aquila-compliant module trains self-organising maps

Libraries

CUDA 5.0
YARP

Parameters

Launch this module with –help parameter to see the list of possible options

Ports Accessed

none

Ports Created

Input ports:

/selfOrganisingMap/[id]:i
- input port to which Aquila connects

Output ports:

/selfOrganisingMap/[id]:o
- output port to which Aquila connects

Input Data Files

Needs a training file, which can be set from:

terminal via the following command: '–input [FILENAME]'
Aquila via the top-level menu

Output Data Files

By default, trained self-organising map is saved to 'som.txt' file. This can be changed from:

terminal via the following command: 'set output [FILENAME]'
Aquila via the top-level menu

Configuration Files

Default values for various paramters are loaded from a configuration file, which can be modified in /conf/config.ini

The file consists in a few sections:

output som.txt
maxThreads 256
learningRate 0.1
sigma 0.0
numSubIterations 100
numOutputs 64

output - output file used for saving trained self-organising map maxThreads 256 - maximum number of GPU threads per block learningRate 0.1 - learning rate used druing the training sigma 0.0 - sigma parameters that changes the training mode if set to a non-zero value numSubIterations - number of sub-iterations numOutputs - number of self-organising map outputs

Tested OS

Linux, OSX and Windows.

Example Instantiation of the Module

This will launch the training on the third GPU on the system (Id=2) for 1000 iterations reading training data from the colour.txt file that can be found in /example folder.

selfOrganisingMap –input colour.txt –iterations 1000 –gpu 2

Author: Martin Peniak Copyright (C) 2008 RobotCub Consortium CopyPolicy: Released under the terms of the BSD This file can be edited at src/main.cpp.