Gaussian Process package based on data augmentation, sparsity and natural gradients
OTHER License
Bot releases are visible (Hide)
Published by github-actions[bot] over 4 years ago
Merged pull requests:
Published by github-actions[bot] over 4 years ago
Closed issues:
Merged pull requests:
Published by github-actions[bot] over 4 years ago
Closed issues:
Merged pull requests:
Published by github-actions[bot] over 4 years ago
Closed issues:
Merged pull requests:
Published by github-actions[bot] over 4 years ago
Closed issues:
Merged pull requests:
Published by julia-tagbot[bot] almost 5 years ago
Large refactoring of the package:
_VGP
and _SVGP
.KernelFunctions.jl
allowing to use automatic differentation, also the derivatives are not adapted for it yet (WIP).AffineMean
has been added to have a parametric mean prior (trainable as well)MOSVGP
a multi-output model has been added to make a linear combination of inducing points see work of Pablo Moreno-MunozMCGP
a sampling based Gaussian Process, for now only Gibbs Sampling is available but it is planned to use HMC via AdvancedHMC.jl
pdf
function (WIP)NumericalVI
is temporarily not working this will be fixed soonMerged pull requests:
Published by julia-tagbot[bot] about 5 years ago
Published by julia-tagbot[bot] over 5 years ago
This release includes
optimizer
and Zoptimizer
Merged pull requests:
Published by theogf over 5 years ago
Corrected Laplace ELBO
Added Poisson Likelihood for events datasets
Custom mean priors : ZeroMean, ConstantMean, EmpiricalMean (see docs)
Published by theogf over 5 years ago
Simplification of the likelihood and inference struct names
Documentation improvement
Published by theogf over 5 years ago
GP
, VGP
and SVGP
parametrized on their likelihood and inferencetrain!(model)
, predict_f(model,X_test)
Makie.plot(model)
rand
Published by theogf almost 6 years ago
Nothing new, just removed the integrated GradDescent
Published by theogf almost 6 years ago
Published by theogf almost 6 years ago
More complete documentation with full examples.
More stable predictions by replacing QuadGK.jl with Expectations.jl (but slightly slower)
Documentation improved for many functions
Published by theogf about 6 years ago
No major improvements
Published by theogf about 6 years ago
Published by theogf about 6 years ago
New major update,
With the new Student-T Likelihood added
Possibility to save and load trained models
Important optimization for the matrix and hyperoptimization side
GradDescent integrated to the package (temporarily)
Published by theogf about 6 years ago
Added a lot of documentation
Better testing
Reconstruction of the structure
First drafts of Multiclass