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

**BREAKING** improve structs & constructors #306

Draft
wants to merge 9 commits into
base: master
Choose a base branch
from
Draft

**BREAKING** improve structs & constructors #306

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
7075809
be more specific about what types work
st-- Mar 30, 2022
f168433
Update src/finite_gp_projection.jl
st-- Mar 30, 2022
ca3cfba
add test
st-- Mar 31, 2022
0ee4a39
add constructor for UniformScaling
st-- Mar 31, 2022
d6fc701
Update test/finite_gp_projection.jl
st-- Mar 31, 2022
1b74c95
Merge branch 'master' into constructors
st-- Mar 31, 2022
c56c412
Merge branch 'master' into constructors
st-- Mar 31, 2022
213b346
protect line
st-- Apr 1, 2022
ac78254
Merge branch 'master' into constructors
st-- Apr 1, 2022
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
9 changes: 7 additions & 2 deletions src/finite_gp_projection.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,8 @@
The finite-dimensional projection of the AbstractGP `f` at `x`. Assumed to be observed under
Gaussian noise with zero mean and covariance matrix `Σy`
"""
struct FiniteGP{Tf<:AbstractGP,Tx<:AbstractVector,TΣ} <: AbstractMvNormal
struct FiniteGP{Tf<:AbstractGP,Tx<:AbstractVector,TΣ<:AbstractMatrix{<:Real}} <:
AbstractMvNormal
f::Tf
x::Tx
Σy::TΣ
Expand All @@ -17,7 +18,11 @@ end
const default_σ² = 1e-18

function FiniteGP(f::AbstractGP, x::AbstractVector, σ²::Real=default_σ²)
return FiniteGP(f, x, Fill(σ², length(x)))
return FiniteGP(f, x, ScalMat(length(x), σ²))
end

function FiniteGP(f::AbstractGP, x::AbstractVector, σ²::UniformScaling)
return FiniteGP(f, x, σ²[1, 1])
end

## conversions
Expand Down
9 changes: 9 additions & 0 deletions test/finite_gp_projection.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -243,6 +243,15 @@ end
first(FiniteDifferences.grad(central_fdm(3, 1), Base.Fix1(logpdf, fx), y))
@test Distributions.sqmahal!(r, fx, Y) ≈ Distributions.sqmahal(fx, Y)
end

@testset "FiniteGP with UniformScaling" begin
f = GP(SqExponentialKernel())
fx = f(rand(10), 2.0 * I)
# for now, just check that it runs
_ = mean(fx)
_ = mean_and_cov(fx)
_ = rand(fx)
end
end

@testset "Docs" begin
Expand Down