Fishtrans object (Multivariate Analysis Toolbox for Matlab®)

This object represents linear transformation using the multidimensional Fisher class-separation criterion. Here, the factors are named discriminant vectors, and their maximal number is g-1, with g being the number of classes.

Navigate to:     General Description     Class Structure     Class Construction     Class Functions

Each field can be accessed by the dot (.) operation, or by the GET function. The GET function can work on multiple instances simultaneously. Most fields, except for those that are Dependent, can be modified using the dot (.) operation, or by the SET function.

		Field	Description	Type	Default	Dedicated Get/Set Function
		ratio	Type of ratio optimized. Can be either 'Sb/Sw' or 'Sb/St'	string	'Sb/Sw'
		ortho_constraints	orthogonality contraints. Each constraint is a column vector.	double matrix	[]
		name	Inherited from lintrans. name of object, should be short and used as identifier. This field will never be empty.	string	'unnamed'
		description	Inherited from lintrans. verbal description of the class content.	string	''
		source	Inherited from lintrans. verbal description of the source of information.	string	''
		type	Inherited from lintrans. type of linear transformation.	string	''
		U	Inherited from lintrans. Transformation matrix.	vvmatrix	[]
		eigvals	Inherited from lintrans. eigenvalues associated with the solution.	double vector	[]
		f_eigvals	Inherited from lintrans. fractional importance of each eigenvalue.	double vector	[]
		no_samples	Inherited from lintrans. number of samples in the original data.	integer	[]	nosamples
		preprocess	Inherited from lintrans. preprocessing of the data matrix. Should be applied to any data before operating with the linear transformation.	vector of preprocess structures	[]
		scores	Inherited from lintrans. the data projected into the new variables (factors).	vsmatrix	[]
	Dependent	no_factors	Inherited from lintrans. number of new factors.	integer scalar	0	nofactors
	Dependent	no_variables	Inherited from lintrans. number of original variables.	integer scalar	0	novariables

Empty instance (scalar)

an empty fishtrans instance, with all fields initialized to their default values.

syntax: ft = fishtrans;

Empty instance (matrix)

an array of empty fishtrans instances.

syntax: ft = fishtrans(size);

Copy constructor

one fishtrans instance is copied into another.

syntax: ft_destination = fishtrans(ft_origin);

Construction by field names

an instance is formed by directly providing field values. Any field which is not dependent is permitted.

syntax: ft = fishtrans(field_name, field_value, ...);

Computing from a vsmatrix

Computing factors from an input in the form of a sample-variable matrix.

syntax: ft = fishtrans(vsm, opt_name, opt_val, ...);

Display functions:

show

displays class content. Source: lintrans.

Operators:

mtimes

operating with the linear transformation Source: lintrans.

SET/GET functions:

eigvals

retrieves the eigenvalues of the linear transformation. Source: lintrans.

factornames

retrieves the names of the factors. Source: lintrans.

factors

retrieves the factors of the linear transformation. Source: lintrans.

get

get method

nofactors

retrieves the number of factors. Source: lintrans.

nosamples

retrieves the number of samples. Source: lintrans.

novariables

retrieves the number of original variables. Source: lintrans.

set

set method

variablenames

retrieves the names of the original variables. Source: lintrans.

Transformations:

deletefactors

excludes certain factors. Source: lintrans.

invertsign

inverts the signs of selected factors. Source: lintrans.

Visualization:

loadings

plots the loadings of the linear transformation. Source: lintrans.

scatter

plots factors versus factors for all data samples Source: lintrans.

scree

plots a scree diagram of the transformation eigenvalues. Source: lintrans.