Click or drag to resize
Accord.NET (logo)

LevenbergMarquardtLearningHessian Property

Gets the approximate Hessian matrix of second derivatives generated in the last algorithm iteration. The Hessian is stored in the upper triangular part of this matrix. See remarks for details.

Namespace:  Accord.Neuro.Learning
Assembly:  Accord.Neuro (in Accord.Neuro.dll) Version: 3.8.0
public float[][] Hessian { get; }
Request Example View Source

Property Value

Type: Single

The Hessian needs only be upper-triangular, since it is symmetric. The Cholesky decomposition will make use of this fact and use the lower-triangular portion to hold the decomposition, conserving memory

Thus said, this property will hold the Hessian matrix in the upper-triangular part of this matrix, and store its Cholesky decomposition on its lower triangular part.

See Also