Packages

c

scalation.analytics

PrincipalComponents

class PrincipalComponents extends Reducer with Error

The PrincipalComponents class performs the Principal Component Analysis 'PCA' on data matrix 'x'. It can be used to reduce the dimensionality of the data. First find the Principal Components 'PC's by calling 'findPCs' and then call 'reduce' to reduce the data (i.e., reduce matrix 'x' to a lower dimensionality matrix).

Linear Supertypes
Error, Reducer, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. PrincipalComponents
  2. Error
  3. Reducer
  4. AnyRef
  5. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new PrincipalComponents(x: MatrixD)

    x

    the data matrix to reduce, stored column-wise

Value Members

  1. def findPCs(k: Int): MatrixD

    Find the Principal Components/Features, the eigenvectors with the 'k' highest eigenvalues.

    Find the Principal Components/Features, the eigenvectors with the 'k' highest eigenvalues.

    k

    the number of Principal Components 'PC's to find

  2. final def flaw(method: String, message: String): Unit
    Definition Classes
    Error
  3. def recover(): MatrixD

    Approximately recover the original data by multiplying the reduced matrix by the inverse (via transpose) of the feature matrix and then adding back the means.

    Approximately recover the original data by multiplying the reduced matrix by the inverse (via transpose) of the feature matrix and then adding back the means.

    Definition Classes
    PrincipalComponents → Reducer
  4. def reduce(): MatrixD

    Multiply the zero mean data matrix by the feature matrix to reduce dimensionality.

    Multiply the zero mean data matrix by the feature matrix to reduce dimensionality.

    Definition Classes
    PrincipalComponents → Reducer