Rename uo --> g0

JuliaEarth · Jan 10, 2025 · 5097c80 · 5097c80
1 parent 1727f8b
commit 5097c80
Show file tree

Hide file tree

Showing 10 changed files with 50 additions and 50 deletions.
diff --git a/src/idw.jl b/src/idw.jl
@@ -52,9 +52,9 @@ end
 # PREDICTION STEP
 #-----------------
 
-predict(fitted::FittedIDW, var, uₒ) = idw(fitted, weights(fitted, uₒ), var)
+predict(fitted::FittedIDW, var, gₒ) = idw(fitted, weights(fitted, gₒ), var)
 
-predictprob(fitted::FittedIDW, var, uₒ) = Dirac(predict(fitted, var, uₒ))
+predictprob(fitted::FittedIDW, var, gₒ) = Dirac(predict(fitted, var, gₒ))
 
 function idw(fitted::FittedIDW, weights, var)
   d = fitted.state.data
@@ -72,16 +72,16 @@ function idw(fitted::FittedIDW, weights, var)
   end
 end
 
-function weights(fitted::FittedIDW, uₒ)
+function weights(fitted::FittedIDW, gₒ)
   e = fitted.model.exponent
   δ = fitted.model.distance
   d = fitted.state.data
   Ω = domain(d)
 
-  # adjust CRS of uₒ
-  uₒ′ = uₒ |> Proj(crs(Ω))
+  # adjust CRS of gₒ
+  gₒ′ = gₒ |> Proj(crs(Ω))
 
-  pₒ = centroid(uₒ′)
+  pₒ = centroid(gₒ′)
   p(i) = centroid(Ω, i)
 
   λ(i) = 1 / evaluate(δ, pₒ, p(i)) ^ e

diff --git a/src/krig/externaldrift.jl b/src/krig/externaldrift.jl
@@ -44,13 +44,13 @@ function set_constraints_lhs!(model::ExternalDriftKriging, LHS::AbstractMatrix,
   nothing
 end
 
-function set_constraints_rhs!(fitted::FittedKriging{<:ExternalDriftKriging}, uₒ)
+function set_constraints_rhs!(fitted::FittedKriging{<:ExternalDriftKriging}, gₒ)
   drifts = fitted.model.drifts
   RHS = fitted.state.RHS
   nobs = nrow(fitted.state.data)
 
   # set external drift
-  pₒ = centroid(uₒ)
+  pₒ = centroid(gₒ)
   for (j, m) in enumerate(drifts)
     RHS[nobs + j] = m(pₒ)
   end

diff --git a/src/krig/ordinary.jl b/src/krig/ordinary.jl
@@ -21,7 +21,7 @@ function set_constraints_lhs!(::OrdinaryKriging, LHS::AbstractMatrix, domain)
   nothing
 end
 
-function set_constraints_rhs!(fitted::FittedKriging{<:OrdinaryKriging}, uₒ)
+function set_constraints_rhs!(fitted::FittedKriging{<:OrdinaryKriging}, gₒ)
   RHS = fitted.state.RHS
   RHS[end] = one(eltype(RHS))
   nothing

diff --git a/src/krig/simple.jl b/src/krig/simple.jl
@@ -28,7 +28,7 @@ nconstraints(::SimpleKriging) = 0
 
 set_constraints_lhs!(::SimpleKriging, LHS::AbstractMatrix, domain) = nothing
 
-set_constraints_rhs!(::FittedKriging{<:SimpleKriging}, uₒ) = nothing
+set_constraints_rhs!(::FittedKriging{<:SimpleKriging}, gₒ) = nothing
 
 function predictmean(fitted::FittedKriging{<:SimpleKriging}, weights::KrigingWeights, var)
   μ = fitted.model.μ

diff --git a/src/krig/universal.jl b/src/krig/universal.jl
@@ -62,14 +62,14 @@ function set_constraints_lhs!(model::UniversalKriging, LHS::AbstractMatrix, doma
   nothing
 end
 
-function set_constraints_rhs!(fitted::FittedKriging{<:UniversalKriging}, uₒ)
+function set_constraints_rhs!(fitted::FittedKriging{<:UniversalKriging}, gₒ)
   exponents = fitted.model.exponents
   RHS = fitted.state.RHS
   nobs = nrow(fitted.state.data)
   nterms = size(exponents, 2)
 
   # set polynomial drift
-  xₒ = CoordRefSystems.raw(coords(centroid(uₒ)))
+  xₒ = CoordRefSystems.raw(coords(centroid(gₒ)))
   for j in 1:nterms
     RHS[nobs + j] = prod(xₒ .^ exponents[:, j])
   end

diff --git a/src/lwr.jl b/src/lwr.jl
@@ -65,46 +65,46 @@ end
 # PREDICTION STEP
 #-----------------
 
-predict(fitted::FittedLWR, var, uₒ) = predictmean(fitted, var, uₒ)
+predict(fitted::FittedLWR, var, gₒ) = predictmean(fitted, var, gₒ)
 
-function predictprob(fitted::FittedLWR, var, uₒ)
-  X, W, A, x, z = matrices(fitted, var, uₒ)
+function predictprob(fitted::FittedLWR, var, gₒ)
+  X, W, A, x, z = matrices(fitted, var, gₒ)
   μ = lwrmean(X, W, A, x, z)
   σ² = lwrvar(X, W, A, x)
   Normal(μ, √σ²)
 end
 
-function predictmean(fitted::FittedLWR, var, uₒ)
-  X, W, A, x, z = matrices(fitted, var, uₒ)
+function predictmean(fitted::FittedLWR, var, gₒ)
+  X, W, A, x, z = matrices(fitted, var, gₒ)
   lwrmean(X, W, A, x, z)
 end
 
-function matrices(fitted::FittedLWR, var, uₒ)
+function matrices(fitted::FittedLWR, var, gₒ)
   d = fitted.state.data
   c = Tables.columns(values(d))
   z = Tables.getcolumn(c, var)
 
-  # adjust CRS of uₒ
-  uₒ′ = uₒ |> Proj(crs(domain(d)))
+  # adjust CRS of gₒ
+  gₒ′ = gₒ |> Proj(crs(domain(d)))
 
   X = fitted.state.X
-  W = wmatrix(fitted, uₒ′)
+  W = wmatrix(fitted, gₒ′)
   A = X' * W * X
 
-  xₒ = CoordRefSystems.raw(coords(centroid(uₒ′)))
+  xₒ = CoordRefSystems.raw(coords(centroid(gₒ′)))
   x = [one(eltype(xₒ)), xₒ...]
 
   X, W, A, x, z
 end
 
-function wmatrix(fitted::FittedLWR, uₒ)
+function wmatrix(fitted::FittedLWR, gₒ)
   w = fitted.model.weightfun
   δ = fitted.model.distance
   d = fitted.state.data
   Ω = domain(d)
   n = nelements(Ω)
 
-  pₒ = centroid(uₒ)
+  pₒ = centroid(gₒ)
   p(i) = centroid(Ω, i)
 
   δs = map(i -> evaluate(δ, pₒ, p(i)), 1:n)

diff --git a/src/models.jl b/src/models.jl
@@ -19,16 +19,16 @@ Fit `model` to geospatial `data` and return a fitted model.
 function fit end
 
 """
-    predict(model, var, uₒ)
+    predict(model, var, gₒ)
 
-Predict variable `var` at geometry `uₒ` with given `model`.
+Predict variable `var` at geometry `gₒ` with given `model`.
 """
 function predict end
 
 """
-    predictprob(model, var, uₒ)
+    predictprob(model, var, gₒ)
 
-Predict distribution of variable `var` at geometry `uₒ` with given `model`.
+Predict distribution of variable `var` at geometry `gₒ` with given `model`.
 """
 function predictprob end
 

diff --git a/src/nn.jl b/src/nn.jl
@@ -40,9 +40,9 @@ end
 # PREDICTION STEP
 #-----------------
 
-predict(fitted::FittedNN, var, uₒ) = nn(fitted, distances(fitted, uₒ), var)
+predict(fitted::FittedNN, var, gₒ) = nn(fitted, distances(fitted, gₒ), var)
 
-predictprob(fitted::FittedNN, var, uₒ) = Dirac(predict(fitted, var, uₒ))
+predictprob(fitted::FittedNN, var, gₒ) = Dirac(predict(fitted, var, gₒ))
 
 function nn(fitted::FittedNN, distances, var)
   d = fitted.state.data
@@ -51,15 +51,15 @@ function nn(fitted::FittedNN, distances, var)
   z[argmin(distances)]
 end
 
-function distances(fitted::FittedNN, uₒ)
+function distances(fitted::FittedNN, gₒ)
   δ = fitted.model.distance
   d = fitted.state.data
   Ω = domain(d)
 
-  # adjust CRS of uₒ
-  uₒ′ = uₒ |> Proj(crs(Ω))
+  # adjust CRS of gₒ
+  gₒ′ = gₒ |> Proj(crs(Ω))
 
-  pₒ = centroid(uₒ′)
+  pₒ = centroid(gₒ′)
   p(i) = centroid(Ω, i)
 
   λ(i) = evaluate(δ, pₒ, p(i))

diff --git a/src/poly.jl b/src/poly.jl
@@ -60,17 +60,17 @@ end
 # PREDICTION STEP
 #-----------------
 
-predict(fitted::FittedPolynomial, var, uₒ) = evalpoly(fitted, var, uₒ)
+predict(fitted::FittedPolynomial, var, gₒ) = evalpoly(fitted, var, gₒ)
 
-predictprob(fitted::FittedPolynomial, var, uₒ) = Dirac(predict(fitted, var, uₒ))
+predictprob(fitted::FittedPolynomial, var, gₒ) = Dirac(predict(fitted, var, gₒ))
 
-function evalpoly(fitted::FittedPolynomial, var, uₒ)
+function evalpoly(fitted::FittedPolynomial, var, gₒ)
   D = domain(fitted.state.data)
   θ = fitted.state.coeffs
   d = fitted.model.degree
-  # adjust CRS of uₒ
-  uₒ′ = uₒ |> Proj(crs(D))
-  xₒ = CoordRefSystems.raw(coords(centroid(uₒ′)))
+  # adjust CRS of gₒ
+  gₒ′ = gₒ |> Proj(crs(D))
+  xₒ = CoordRefSystems.raw(coords(centroid(gₒ′)))
   V = vandermonde((xₒ,), d)
   first(V * θ[var])
 end

diff --git a/test/krig.jl b/test/krig.jl
@@ -236,11 +236,11 @@
     dk = GeoStatsModels.fit(DK(γ, [x -> 1.0]), data)
 
     # predict on a quadrangle
-    uₒ = Quadrangle((0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0))
-    skdist = GeoStatsModels.predictprob(sk, :z, uₒ)
-    okdist = GeoStatsModels.predictprob(ok, :z, uₒ)
-    ukdist = GeoStatsModels.predictprob(uk, :z, uₒ)
-    dkdist = GeoStatsModels.predictprob(dk, :z, uₒ)
+    gₒ = Quadrangle((0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0))
+    skdist = GeoStatsModels.predictprob(sk, :z, gₒ)
+    okdist = GeoStatsModels.predictprob(ok, :z, gₒ)
+    ukdist = GeoStatsModels.predictprob(uk, :z, gₒ)
+    dkdist = GeoStatsModels.predictprob(dk, :z, gₒ)
 
     # variance checks
     @test var(skdist) ≥ 0
@@ -266,11 +266,11 @@
     dk = GeoStatsModels.fit(DK(γ, [x -> 1.0]), data)
 
     # prediction on a quadrangle
-    uₒ = Quadrangle((0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0))
-    skmean = GeoStatsModels.predict(sk, :z, uₒ)
-    okmean = GeoStatsModels.predict(ok, :z, uₒ)
-    ukmean = GeoStatsModels.predict(uk, :z, uₒ)
-    dkmean = GeoStatsModels.predict(dk, :z, uₒ)
+    gₒ = Quadrangle((0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0))
+    skmean = GeoStatsModels.predict(sk, :z, gₒ)
+    okmean = GeoStatsModels.predict(ok, :z, gₒ)
+    ukmean = GeoStatsModels.predict(uk, :z, gₒ)
+    dkmean = GeoStatsModels.predict(dk, :z, gₒ)
 
     # type tests
     @test skmean isa Composition