The purpose of this PR is to to be able to use TrajectorySamplers within Trieste (further on HV repo) that can support Heteroskedastic likelihoods (so multiple GP heads) that can have SeparateIndependent and SharedIndependent kernels & inducing variables.
Main changes:
creation of feature_decomposition_kernels folder. Idea was to structure it just like in GPflow (i.e. multioutput subfolder). Having a separate folder for this type of kernels will prove to be better suited if we are planning on including in the future some other papers such as .. [1] Solin, Arno, and Simo Särkkä. "Hilbert space methods for reduced-rank Gaussian process regression." Statistics and Computing (2020). .. [2] Borovitskiy, Viacheslav, et al. "Matérn Gaussian processes on Riemannian manifolds." In Advances in Neural Information Processing Systems (2020).
in gpflux.layers.basis_functions.fourier_features I have added the multioutput version
in gpflux.sampling I have added the multioutput version
The purpose of this PR is to to be able to use TrajectorySamplers within Trieste (further on HV repo) that can support Heteroskedastic likelihoods (so multiple GP heads) that can have SeparateIndependent and SharedIndependent kernels & inducing variables.
Main changes: