Skip to content

Releases: iqiukp/SVDD-MATLAB

v2.2

12 May 17:38
f526a9c
Compare
Choose a tag to compare
  • Added support for hybrid-kernel SVDD.
    K =w1K1+w2K2+...+wn*Kn
  • Added support for output of optimization results.
  • Improved visualization of parameter optimization.
  • Fixed minor bugs.

v2.1.5

25 Jan 03:15
17d1444
Compare
Choose a tag to compare
Update README.md

v2.1.2

24 Jan 13:50
Compare
Choose a tag to compare
Update demo_nSVDD.m

v2.1.1

21 Dec 17:59
7cc048d
Compare
Choose a tag to compare
  • Fixed bugs with Display Settings for Parameter Optimization.

v2.1

11 May 05:00
65e90d3
Compare
Choose a tag to compare
  1. SVDD model for one-class or binary classification
  2. Multiple kinds of kernel functions (linear, gaussian, polynomial, sigmoid, laplacian)
  3. Visualization of decision boundaries for 2D or 3D data
  4. Parameter Optimization using Bayesian optimization, Genetic Algorithm, and Particle Swarm Optimization
  5. Weighted SVDD model

v2.0

01 Dec 05:17
5b9ecf9
Compare
Choose a tag to compare
  1. SVDD model for training dataset containing only positive training data. (SVDD)
  2. SVDD model for training dataset containing both positive training data and negative training data. (nSVDD)
  3. Multiple kinds of kernel functions.
  4. Visualization module including ROC curve plotting, testing result plotting, and decision boundary.
  5. Dimensionality Reduction (DR) module based on 'drtoolbox'.
  6. Parameter Optimization (PO) module based on the 'YPEA' toolbox.
  7. Preprocessing module for data standardization or data normalization.

v1.0

30 Nov 06:48
c97d160
Compare
Choose a tag to compare
Update README.md