Skip to content
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

Jump to bottom

RFC: Refactor LatentGP API slightly #216

Draft
wants to merge 3 commits into
base: master
Choose a base branch
from
Draft

RFC: Refactor LatentGP API slightly #216

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2453565
Tweak LatentGP API
willtebbutt Aug 25, 2021
1bb6e7e
Bump minor version to DEV
willtebbutt Aug 25, 2021
cd59bf7
Update docstring
willtebbutt Aug 25, 2021
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
@@ -1,7 +1,7 @@
name = "AbstractGPs"
uuid = "99985d1d-32ba-4be9-9821-2ec096f28918"
authors = ["JuliaGaussianProcesses Team"]
version = "0.4.0"
version = "0.5.0-DEV"

[deps]
ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"
Expand Down
8 changes: 5 additions & 3 deletions src/latent_gp.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,10 @@
LatentGP(f<:GP, lik, Σy)

- `f` is a `AbstractGP`.
- `lik` is the likelihood function which maps samples from `f` to the corresponding
conditional likelihood distributions (i.e., `lik` must return a `Distribution` compatible with the observations).
- `lik` is a function mapping inputs `x` to a likelihood function, itself mapping samples
from `f` to the corresponding conditional likelihood distributions (i.e., `lik` must
return another function, which returns a `Distribution` compatible with the observations
when called at values of the latent process).
- `Σy` is the noise under which the latent GP is "observed"; this represents the jitter used to avoid numeric instability and should generally be small.

"""
Expand All @@ -26,7 +28,7 @@ struct LatentFiniteGP{Tfx<:FiniteGP,Tlik}
lik::Tlik
end

(lgp::LatentGP)(x) = LatentFiniteGP(lgp.f(x, lgp.Σy), lgp.lik)
(lgp::LatentGP)(x) = LatentFiniteGP(lgp.f(x, lgp.Σy), lgp.lik(x))

function Distributions.rand(rng::AbstractRNG, lfgp::LatentFiniteGP)
f = rand(rng, lfgp.fx)
Expand Down
3 changes: 1 addition & 2 deletions test/latent_gp.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,10 +3,9 @@
x = rand(10)
y = rand(10)

lgp = LatentGP(gp, x -> MvNormal(x, 0.1), 1e-5)
lgp = LatentGP(gp, x -> (f -> MvNormal(f, 0.1)), 1e-5)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This constructor of MvNormal was deprecated (JuliaStats/Distributions.jl#1362), it should be replaced with e.g.

Suggested change
lgp = LatentGP(gp, x -> (f -> MvNormal(f, 0.1)), 1e-5)
lgp = LatentGP(gp, x -> (f -> MvNormal(f, 0.01 * I)), 1e-5)

@test lgp isa LatentGP
@test lgp.f isa AbstractGPs.AbstractGP
@test lgp.Σy isa Real

lfgp = lgp(x)
@test lfgp isa AbstractGPs.LatentFiniteGP
Expand Down