Skip to content

build on approx_log_evidence introduced in AbstractGPs 0.5.17 #145

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

build on approx_log_evidence introduced in AbstractGPs 0.5.17 #145

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
e7f4b36
build on approx_log_evidence introduced in AbstractGPs 0.5.17
st-- May 8, 2023
a4734bb
Merge branch 'master' into st/3posterior
st-- May 8, 2023
804b2fa
update tests
st-- May 8, 2023
9000c74
fix
st-- May 9, 2023
3f9b152
bump version
st-- May 9, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"

[compat]
AbstractGPs = "0.3, 0.4, 0.5"
AbstractGPs = "0.5.17"
ChainRulesCore = "1.7"
Distributions = "0.25"
FastGaussQuadrature = "0.4, 0.5"
Expand Down
10 changes: 1 addition & 9 deletions src/API.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,14 +2,6 @@ module API

export approx_lml # TODO move to AbstractGPs, see https://github.com/JuliaGaussianProcesses/AbstractGPs.jl/issues/221

"""
approx_lml(approx::<Approximation>, lfx::LatentFiniteGP, ys)

Compute an approximation to the log of the marginal likelihood (also known as
"evidence") under the given `approx` to the posterior. This approximation can be used to optimise the hyperparameters of `lfx`.

This should become part of the AbstractGPs API (see JuliaGaussianProcesses/AbstractGPs.jl#221).
"""
function approx_lml end
@deprecate approx_lml(approx, lfx::LatentFiniteGP, ys) approx_log_evidence(approx, lfx, ys)

end
4 changes: 2 additions & 2 deletions src/LaplaceApproximationModule.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,14 +48,14 @@ function AbstractGPs.posterior(la::LaplaceApproximation, lfx::LatentFiniteGP, ys
end

"""
approx_lml(la::LaplaceApproximation, lfx::LatentFiniteGP, ys)
approx_log_evidence(la::LaplaceApproximation, lfx::LatentFiniteGP, ys)

Compute an approximation to the log of the marginal likelihood (also known as
"evidence"), which can be used to optimise the hyperparameters of `lfx`.

This should become part of the AbstractGPs API (see JuliaGaussianProcesses/AbstractGPs.jl#221).
"""
function API.approx_lml(la::LaplaceApproximation, lfx::LatentFiniteGP, ys)
function AbstractGPs.approx_log_evidence(la::LaplaceApproximation, lfx::LatentFiniteGP, ys)
return laplace_lml(lfx, ys; la.newton_kwargs...)
end

Expand Down
4 changes: 2 additions & 2 deletions src/SparseVariationalApproximationModule.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -273,10 +273,10 @@ inducing_points(f::ApproxPosteriorGP{<:SparseVariationalApproximation}) = f.appr
# elbo
#

function API.approx_lml(
function AbstractGPs.approx_log_evidence(
sva::SparseVariationalApproximation, l_fx::Union{FiniteGP,LatentFiniteGP}, ys; kwargs...
)
return AbstractGPs.elbo(sva, l_fx, ys; kwargs...)
return elbo(sva, l_fx, ys; kwargs...)
end

"""
Expand Down
4 changes: 2 additions & 2 deletions src/TestUtils.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ function build_latent_gp(theta)
end

"""
test_approx_lml(approx)
test_approx_log_evidence(approx)

Test whether in the conjugate case `approx_lml(approx, LatentGP(f,
GaussianLikelihood(), jitter)(x), y)` gives approximately the same answer as
Expand All @@ -51,7 +51,7 @@ the log marginal likelihood in exact GP regression.
!!! todo
Also test gradients (for hyperparameter optimization).
"""
function test_approx_lml end
function test_approx_log_evidence end

"""
test_approximation_predictions(approx)
Expand Down