OpenANN  1.1.0
An open source library for artificial neural networks.
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Pages
List of all members | Public Member Functions
OpenANN::Compressor Class Reference

Compresses arbitrary one-dimensional data. More...

#include <Compressor.h>

+ Inheritance diagram for OpenANN::Compressor:

Public Member Functions

 Compressor (int inputDim, int outputDim, CompressionMatrixFactory::Transformation transformation)
 
virtual Transformerfit (const Eigen::MatrixXd &X)
 Fit transformation according to training set X. More...
 
virtual TransformerfitPartial (const Eigen::MatrixXd &X)
 Fit transformation according to subset of the training set X. More...
 
virtual Eigen::MatrixXd transform (const Eigen::MatrixXd &X)
 Transform the data. More...
 
int getOutputs ()
 
- Public Member Functions inherited from OpenANN::Transformer
virtual ~Transformer ()
 

Detailed Description

Compresses arbitrary one-dimensional data.

The compression must have the form $ \Phi x = y $, where $ \Phi $ can be an arbitrary compression matrix, $ x $ is the D-dimensional input and $ y $ is the F-dimensional output (i.e. F < D).

Constructor & Destructor Documentation

OpenANN::Compressor::Compressor ( int  inputDim,
int  outputDim,
CompressionMatrixFactory::Transformation  transformation 
)

Member Function Documentation

virtual Transformer& OpenANN::Compressor::fit ( const Eigen::MatrixXd &  X)
virtual

Fit transformation according to training set X.

Parameters
Xeach row represents an instance
Returns
this for chaining

Implements OpenANN::Transformer.

virtual Transformer& OpenANN::Compressor::fitPartial ( const Eigen::MatrixXd &  X)
virtual

Fit transformation according to subset of the training set X.

This can be used for online adaption of the transformation.

Parameters
Xeach row represents an instance
Returns
this for chaining

Reimplemented from OpenANN::Transformer.

int OpenANN::Compressor::getOutputs ( )
virtual Eigen::MatrixXd OpenANN::Compressor::transform ( const Eigen::MatrixXd &  X)
virtual

Transform the data.

Parameters
Xeach row represents an instance
Returns
transformed data

Implements OpenANN::Transformer.


The documentation for this class was generated from the following file: