Releases: yuki-koyama/mathtoolbox
Releases · yuki-koyama/mathtoolbox
v0.3
This release includes many changes. Major changes are as follows.
New modules:
som
: Self-organizing map (SOM)backtracking-line-search
: Backtracking line searchbayesian-optimization
: Bayesian optimizationbfgs
: BFGS methodgradient-descent
: Gradient descent methodl-bfgs
: Limited-memory BFGS methodstrong-wolfe-conditions-line-search
: Strong Wolfe conditions line searchlog-determinant
: Log-determinantmatrix-inversion
: Matrix inversion techniquesacquisition-functions
: Acquisition functionsconstants
: Constantsdata-normalization
: Data normalizationkernel-functions
: Kernel functionsprobability-distributions
: Probability distributions
New features:
- Python bindings
Test environments:
- GitHub Actions
v0.2.0
This release includes the following important changes:
- Add support of Gaussian process regression
v0.1.0
Add the following algorithms:
- rbf-interpolation
- classical-mds