FFNtrainInfo Class Reference

Collaboration diagram for FFNtrainInfo:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 FFNtrainInfo (std::string fName=std::string(""))
 Constructor.
virtual ~FFNtrainInfo ()
 Destructor.
bool reset (std::string fName)
 reset the Neural Network classifier parameter

Public Attributes

std::string trainFolder
 where the training data is
std::string testFolder
 where the testing data is
uint nOutput
 the number of output dimension
bool isPCA
 has a dimension reduction step
std::string evecFname
 the dimension reduction
uint oriFeatSize
 the original number of features
uint redFeatSize
 the reduced number of features
uint h1size
 the number of hidden nodes in layer 1
uint h2size
 the number of hidden nodes in layer 2
std::string h1Name
 the weight file for hidden layer 1
std::string h2Name
 the weight file for hidden layer 2
std::string oName
 the weight file for output layer
float learnRate
 the learning rate
std::string trainSampleFile
 the training sample file list
std::string testSampleFile
 the testing sample file list

Detailed Description

Definition at line 53 of file trainUtils.H.


Constructor & Destructor Documentation

FFNtrainInfo::FFNtrainInfo ( std::string  fName = std::string("")  ) 

Constructor.

Construct a FFN training params info if blank, need to call reset later on

Definition at line 48 of file trainUtils.C.

References reset().

FFNtrainInfo::~FFNtrainInfo (  )  [virtual]

Destructor.

Definition at line 55 of file trainUtils.C.


Member Function Documentation

bool FFNtrainInfo::reset ( std::string  fName  ) 

reset the Neural Network classifier parameter

reset the training info with a new file

Definition at line 60 of file trainUtils.C.

References evecFname, h1Name, h1size, h2Name, h2size, isPCA, learnRate, nOutput, oName, oriFeatSize, redFeatSize, testFolder, testSampleFile, trainFolder, and trainSampleFile.

Referenced by FFNtrainInfo().


Member Data Documentation

the dimension reduction

Definition at line 70 of file trainUtils.H.

Referenced by reset().

the weight file for hidden layer 1

Definition at line 75 of file trainUtils.H.

Referenced by reset().

the number of hidden nodes in layer 1

Definition at line 73 of file trainUtils.H.

Referenced by reset().

the weight file for hidden layer 2

Definition at line 76 of file trainUtils.H.

Referenced by reset().

the number of hidden nodes in layer 2

Definition at line 74 of file trainUtils.H.

Referenced by reset().

has a dimension reduction step

Definition at line 69 of file trainUtils.H.

Referenced by Environment::classifySegNum(), and reset().

the learning rate

Definition at line 78 of file trainUtils.H.

Referenced by reset().

the number of output dimension

Definition at line 68 of file trainUtils.H.

Referenced by reset().

the weight file for output layer

Definition at line 77 of file trainUtils.H.

Referenced by reset().

the original number of features

Definition at line 71 of file trainUtils.H.

Referenced by reset().

the reduced number of features

Definition at line 72 of file trainUtils.H.

Referenced by reset().

where the testing data is

Definition at line 67 of file trainUtils.H.

Referenced by reset().

the testing sample file list

Definition at line 80 of file trainUtils.H.

Referenced by reset().

where the training data is

Definition at line 66 of file trainUtils.H.

Referenced by reset().

the training sample file list

Definition at line 79 of file trainUtils.H.

Referenced by reset().


The documentation for this class was generated from the following files:
Generated on Sun May 8 08:43:20 2011 for iLab Neuromorphic Vision Toolkit by  doxygen 1.6.3