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

Search methods: handle different units #1064

Merged
merged 3 commits into from
Sep 16, 2024
Merged

Search methods: handle different units #1064

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
c4644ed
Search methods: handle different units
eliascarv Sep 16, 2024
d7986ca
Minor adjustments
juliohm Sep 16, 2024
b662c8c
Minor adjustments
juliohm Sep 16, 2024
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
8 changes: 6 additions & 2 deletions src/neighborsearch/ball.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,9 +27,13 @@ BallSearch(geoms, ball) = BallSearch(GeometrySet(geoms), ball)

function search(pₒ::Point, method::BallSearch; mask=nothing)
tree = method.tree
dmax = radius(method.ball)

inds = inrange(tree, ustrip.(to(pₒ)), ustrip(dmax))
# adjust units of query point and radius
u = unit(lentype(method.domain))
r = ustrip(u, radius(method.ball))
x = ustrip.(u, to(pₒ))

inds = inrange(tree, x, r)

if isnothing(mask)
inds
Expand Down
11 changes: 7 additions & 4 deletions src/neighborsearch/kball.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,15 +30,18 @@ KBallSearch(geoms, k, ball) = KBallSearch(GeometrySet(geoms), k, ball)
maxneighbors(method::KBallSearch) = method.k

function searchdists!(neighbors, distances, pₒ::Point, method::KBallSearch; mask=nothing)
u = unit(lentype(pₒ))
tree = method.tree
dmax = radius(method.ball)
k = method.k

inds, dists = knn(tree, ustrip.(to(pₒ)), k, true)
# adjust units of query point and radius
u = unit(lentype(method.domain))
r = ustrip(u, radius(method.ball))
x = ustrip.(u, to(pₒ))

inds, dists = knn(tree, x, k, true)

# keep neighbors inside ball
keep = dists .≤ ustrip(dmax)
keep = dists .≤ r

# possibly mask some of the neighbors
isnothing(mask) || (keep .*= mask[inds])
Expand Down
7 changes: 5 additions & 2 deletions src/neighborsearch/knearest.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,11 +28,14 @@ KNearestSearch(geoms, k; metric=Euclidean()) = KNearestSearch(GeometrySet(geoms)
maxneighbors(method::KNearestSearch) = method.k

function searchdists!(neighbors, distances, pₒ::Point, method::KNearestSearch; mask=nothing)
u = unit(lentype(pₒ))
tree = method.tree
k = method.k

inds, dists = knn(tree, ustrip.(to(pₒ)), k, true)
# adjust units of query point
u = unit(lentype(method.domain))
x = ustrip.(u, to(pₒ))

inds, dists = knn(tree, x, k, true)

if isnothing(mask)
nneigh = k
Expand Down
21 changes: 21 additions & 0 deletions test/neighborsearch.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,13 @@
n = search(cart(9, 9), s)
@test Set(n) == Set([89, 90, 99, 100])

# different units
s = BallSearch(𝒟, MetricBall(T(10) * u"dm"))
n = search(Point(T(900) * u"cm", T(900) * u"cm"), s)
@test Set(n) == Set([100, 99, 90])
n = search(Point(T(9000) * u"mm", T(9000) * u"mm"), s)
@test Set(n) == Set([100, 99, 90])

# non MinkowskiMetric example
𝒟 = CartesianGrid((360, 180), T.((0.0, -90.0)), T.((1.0, 1.0)))
s = BallSearch(𝒟, MetricBall(T(150), Haversine(T(6371))))
Expand Down Expand Up @@ -56,6 +63,13 @@ end
@test nn == 3
@test Set(n[1:nn]) == Set([100, 99, 90])

# different units
s = KNearestSearch(𝒟, 3)
n = search(Point(T(900) * u"cm", T(900) * u"cm"), s)
@test Set(n) == Set([100, 99, 90])
n = search(Point(T(9000) * u"mm", T(9000) * u"mm"), s)
@test Set(n) == Set([100, 99, 90])

# construct from vector of geometries
s = KNearestSearch(randpoint2(100), 3)
@test s isa KNearestSearch
Expand Down Expand Up @@ -105,6 +119,13 @@ end
@test length(n) == 4
@test length(d) == 4

# different units
s = KBallSearch(𝒟, 10, MetricBall(T(10) * u"dm"))
n = search(Point(T(500) * u"cm", T(500) * u"cm"), s)
@test Set(n) == Set([56, 66, 55, 57, 46])
n = search(Point(T(5000) * u"mm", T(5000) * u"mm"), s)
@test Set(n) == Set([56, 66, 55, 57, 46])

# construct from vector of geometries
s = KBallSearch(randpoint2(100), 10, MetricBall(T(1)))
@test s isa KBallSearch
Expand Down
Loading