SciML · ChrisRackauckas · Aug 5, 2025 · Jul 23, 2025 · Jul 23, 2025 · Jul 23, 2025
diff --git a/.github/workflows/CompatHelper.yml b/.github/workflows/CompatHelper.yml
@@ -24,10 +24,4 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: >
-          julia -e 'using CompatHelper; CompatHelper.main(;subdirs=["", "docs",
-                                                                    "lib/SurrogatesAbstractGPs",
-                                                                    "lib/SurrogatesFlux",
-                                                                    "lib/SurrogatesMOE",
-                                                                    "lib/SurrogatesPolyChaos",
-                                                                    "lib/SurrogatesRandomForest",
-                                                                    "lib/SurrogatesSVM"])'
+          julia -e 'using CompatHelper; CompatHelper.main(;subdirs=["", "docs"])'
diff --git a/.github/workflows/Documentation.yml b/.github/workflows/Documentation.yml
@@ -18,12 +18,6 @@ jobs:
       - name: Install dependencies
         run: julia --project=docs/ -e 'using Pkg; 
                                       Pkg.develop(PackageSpec(path=pwd()));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesAbstractGPs")));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesFlux")));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesPolyChaos")));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesRandomForest")));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesMOE")));
-                                      Pkg.develop(PackageSpec(path=joinpath(pwd(), "lib", "SurrogatesSVM")));
                                       Pkg.instantiate()'
       - name: Build and deploy
         env:
@@ -32,7 +26,7 @@ jobs:
         run: julia --project=docs/ --code-coverage=user docs/make.jl
       - uses: julia-actions/julia-processcoverage@v1
         with:
-          directories: src,lib/SurrogatesAbstractGPs/src,lib/SurrogatesFlux/src,lib/SurrogatesMOE/src,lib/SurrogatesPolyChaos/src,lib/SurrogatesRandomForest/src,lib/SurrogatesSVM/src
+          directories: src,ext
       - uses: codecov/codecov-action@v5
         with:
           file: lcov.info

diff --git a/.github/workflows/Downgrade.yml b/.github/workflows/Downgrade.yml
@@ -15,8 +15,6 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        group:
-          - Core
         downgrade_mode: ['alldeps']
         julia-version: ['1.10']
     steps:
@@ -25,12 +23,9 @@ jobs:
         with:
           version: ${{ matrix.julia-version }}
       - uses: julia-actions/julia-downgrade-compat@v2
-#        if: ${{ matrix.version == '1.6' }}
         with:
           skip: Pkg,TOML
       - uses: julia-actions/julia-buildpkg@v1
       - uses: julia-actions/julia-runtest@v1
         with:
           ALLOW_RERESOLVE: false
-        env:
-          GROUP: ${{ matrix.group }}
diff --git a/.github/workflows/Tests.yml b/.github/workflows/Tests.yml
@@ -33,5 +33,5 @@ jobs:
     with:
       julia-version: "${{ matrix.version }}"
       group: "${{ matrix.group }}"
-      coverage-directories: "src,lib/SurrogatesAbstractGPs/src,lib/SurrogatesFlux/src,lib/SurrogatesMOE/src,lib/SurrogatesPolyChaos/src,lib/SurrogatesRandomForest/src,lib/SurrogatesSVM/src"
+      coverage-directories: "src,ext"
     secrets: "inherit"
diff --git a/Project.toml b/Project.toml
@@ -11,44 +11,76 @@ GLM = "38e38edf-8417-5370-95a0-9cbb8c7f171a"
 IterativeSolvers = "42fd0dbc-a981-5370-80f2-aaf504508153"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 QuasiMonteCarlo = "8a4e6c94-4038-4cdc-81c3-7e6ffdb2a71b"
+Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 SurrogatesBase = "89f642e6-4179-4274-8202-c11f4bd9a72c"
 Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
+[weakdeps]
+AbstractGPs = "99985d1d-32ba-4be9-9821-2ec096f28918"
+Flux = "587475ba-b771-5e3f-ad9e-33799f191a9c"
+GaussianMixtures = "cc18c42c-b769-54ff-9e2a-b28141a64aae"
+LIBSVM = "b1bec4e5-fd48-53fe-b0cb-9723c09d164b"
+Optimisers = "3bd65402-5787-11e9-1adc-39752487f4e2"
+PolyChaos = "8d666b04-775d-5f6e-b778-5ac7c70f65a3"
+XGBoost = "009559a3-9522-5dbb-924b-0b6ed2b22bb9"
+
+[extensions]
+SurrogatesAbstractGPsExt = "AbstractGPs"
+SurrogatesFluxExt = ["Flux"]
+SurrogatesPolyChaosExt = "PolyChaos"
+SurrogatesXGBoostExt = "XGBoost"
+SurrogatesSVMExt = "LIBSVM"
+SurrogatesMOEExt = "GaussianMixtures"
+
 [compat]
+AbstractGPs = "0.5"
 Aqua = "0.8"
 ChainRulesCore = "1.19.1"
 Cubature = "1.5"
-Distributions = "0.25.71"
+Distributions = "0.25.120"
 ExtendableSparse = "1"
 Flux = "0.15, 0.16"
-ForwardDiff = "0.10.19"
-GLM = "1.5"
+ForwardDiff = "0.10.38"
+GaussianMixtures = "0.3.13"
+GLM = "1.9"
 IterativeSolvers = "0.9"
+LIBSVM = "0.8"
 LinearAlgebra = "1.10"
+Optimisers = "0.4.4"
 Pkg = "1.10"
 PolyChaos = "0.2.5"
 QuadGK = "2.4"
-QuasiMonteCarlo = "0.3.1"
+QuasiMonteCarlo = "0.3.3"
+Random = "1.10"
 SafeTestsets = "0.1"
 SparseArrays = "1.10"
+StableRNGs = "1"
 Statistics = "1.10"
 SurrogatesBase = "1.1.0"
 Test = "1.10"
+XGBoost = "2"
 Zygote = "0.7"
 julia = "1.10"
 
 [extras]
+AbstractGPs = "99985d1d-32ba-4be9-9821-2ec096f28918"
 Aqua = "4c88cf16-eb10-579e-8560-4a9242c79595"
 Cubature = "667455a9-e2ce-5579-9412-b964f529a492"
 Flux = "587475ba-b771-5e3f-ad9e-33799f191a9c"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
+GaussianMixtures = "cc18c42c-b769-54ff-9e2a-b28141a64aae"
+LIBSVM = "b1bec4e5-fd48-53fe-b0cb-9723c09d164b"
+Optimisers = "3bd65402-5787-11e9-1adc-39752487f4e2"
 Pkg = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
 PolyChaos = "8d666b04-775d-5f6e-b778-5ac7c70f65a3"
 QuadGK = "1fd47b50-473d-5c70-9696-f719f8f3bcdc"
+Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 SafeTestsets = "1bc83da4-3b8d-516f-aca4-4fe02f6d838f"
+StableRNGs = "860ef19b-820b-49d6-a774-d7a799459cd3"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+XGBoost = "009559a3-9522-5dbb-924b-0b6ed2b22bb9"
 
 [targets]
-test = ["Aqua", "Cubature", "SafeTestsets", "Flux", "ForwardDiff", "PolyChaos", "QuadGK", "Test", "Pkg"]
+test = ["AbstractGPs", "Aqua", "Cubature", "SafeTestsets", "Flux", "ForwardDiff", "GaussianMixtures", "LIBSVM", "Optimisers", "PolyChaos", "QuadGK", "Random", "StableRNGs", "Test", "Pkg", "XGBoost"]
diff --git a/docs/Project.toml b/docs/Project.toml
@@ -2,30 +2,22 @@
 AbstractGPs = "99985d1d-32ba-4be9-9821-2ec096f28918"
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
 Flux = "587475ba-b771-5e3f-ad9e-33799f191a9c"
+GaussianMixtures = "cc18c42c-b769-54ff-9e2a-b28141a64aae"
+PolyChaos = "8d666b04-775d-5f6e-b778-5ac7c70f65a3"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 QuadGK = "1fd47b50-473d-5c70-9696-f719f8f3bcdc"
 Surrogates = "6fc51010-71bc-11e9-0e15-a3fcc6593c49"
-SurrogatesAbstractGPs = "78aa1720-c2af-471b-b307-964fd38f9b5f"
-SurrogatesFlux = "4f55584b-dac4-4b1c-802a-b7c47b72cb4c"
-SurrogatesMOE = "778709c9-3595-4497-80d4-d1077d5b9fbf"
-SurrogatesPolyChaos = "50679fc6-c85c-4a6e-ac63-dc3c8bd8cb1c"
-SurrogatesRandomForest = "3fee2672-df33-422b-aa65-d915eeac013a"
-SurrogatesSVM = "eb0c118b-dd0a-445a-b996-82088348ceb1"
 XGBoost = "009559a3-9522-5dbb-924b-0b6ed2b22bb9"
 Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
 [compat]
 AbstractGPs = "0.5.13"
 Documenter = "1"
 Flux = "0.15, 0.16"
+GaussianMixtures = "0.3.6"
+PolyChaos = "0.2.5"
 Plots = "1.36.2"
 QuadGK = "2.6.0"
 Surrogates = "6"
-SurrogatesAbstractGPs = "0.1.0"
-SurrogatesFlux = "0.1.0"
-SurrogatesMOE = "0.1.0"
-SurrogatesPolyChaos = "0.1.0"
-SurrogatesRandomForest = "0.1.0"
-SurrogatesSVM = "0.1.0"
 XGBoost = "2"
 Zygote = "0.6.49, 0.7"
diff --git a/docs/pages.jl b/docs/pages.jl
@@ -7,7 +7,7 @@ pages = ["index.md"
              "Lobachevsky" => "lobachevsky.md",
              "Linear" => "LinearSurrogate.md",
              "InverseDistance" => "InverseDistance.md",
-             "RandomForest" => "randomforest.md",
+             "XGBoost" => "xgboost.md",
              "SecondOrderPolynomial" => "secondorderpoly.md",
              "NeuralSurrogate" => "neural.md",
              "Wendland" => "wendland.md",

diff --git a/docs/src/abstractgps.md b/docs/src/abstractgps.md
@@ -1,9 +1,5 @@
 # Gaussian Process Surrogate Tutorial
 
-!!! note
-
-    This surrogate requires the 'SurrogatesAbstractGPs' module, which can be added by inputting "]add SurrogatesAbstractGPs" from the Julia command line.
-
 Gaussian Process regression in Surrogates.jl is implemented as a simple wrapper around the [AbstractGPs.jl](https://github.com/JuliaGaussianProcesses/AbstractGPs.jl) package. AbstractGPs comes with a variety of covariance functions (kernels). See [KernelFunctions.jl](https://github.com/JuliaGaussianProcesses/KernelFunctions.jl/) for examples.
 
 !!! tip
@@ -19,8 +15,7 @@ In the example below, the 'gp_surrogate' assignment code can be commented / unco
 ```@example gp_tutorial1d
 using Surrogates
 using Plots
-using AbstractGPs #required to access different types of kernels
-using SurrogatesAbstractGPs
+using AbstractGPs
 
 f(x) = (6 * x - 2)^2 * sin(12 * x - 4)
 n_samples = 100
@@ -46,7 +41,6 @@ This example shows the use of AbstractGP Surrogates to find the minima of a func
 using Surrogates
 using Plots
 using AbstractGPs
-using SurrogatesAbstractGPs
 
 f(x) = (x - 2)^2
 n_samples = 100
@@ -65,7 +59,7 @@ Plotting the function and the sampled points:
 scatter(gp_surrogate.x, gp_surrogate.y, label = "Sampled points",
     ylims = (-1.0, 5.0), legend = :top)
 plot!(xs, gp_surrogate.(xs), label = "Surrogate function",
-    ribbon = p -> SurrogatesAbstractGPs.std_error_at_point(gp_surrogate, p), legend = :top)
+    ribbon = p -> Surrogates.std_error_at_point(gp_surrogate, p), legend = :top)
 plot!(xs, f.(xs), label = "True function", legend = :top)
 ```
 
@@ -76,7 +70,6 @@ using Plots
 default(c = :matter, legend = false, xlabel = "x", ylabel = "y")
 using Surrogates
 using AbstractGPs
-using SurrogatesAbstractGPs
 
 hypot_func = z -> 3 * hypot(z...) + 1
 n_samples = 100

diff --git a/docs/src/cantilever.md b/docs/src/cantilever.md
@@ -8,7 +8,7 @@ Let's import Surrogates and Plots:
 
 ```@example beam
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 ```
 

diff --git a/docs/src/gramacylee.md b/docs/src/gramacylee.md
@@ -10,7 +10,7 @@ Let's import these two packages `Surrogates` and `Plots`:
 
 ```@example gramacylee1D
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 ```
 

diff --git a/docs/src/lp.md b/docs/src/lp.md
@@ -7,7 +7,7 @@ Let's import Surrogates and Plots:
 
 ```@example lp
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 using LinearAlgebra
 ```

diff --git a/docs/src/moe.md b/docs/src/moe.md
@@ -1,9 +1,5 @@
 # Mixture of Experts (MOE) Surrogate Tutorial
 
-!!! note
-
-    This surrogate requires the 'SurrogatesMOE' module, which can be added by inputting "]add SurrogatesMOE" from the Julia command line.
-
 The Mixture of Experts (MOE) Surrogate model represents the interpolating function as a combination of other surrogate models. SurrogatesMOE is a Julia implementation of the [Python version from SMT](https://smt.readthedocs.io/en/latest/_src_docs/applications/moe.html).
 
 MOE is most useful when we have a discontinuous function. For example, let's say we want to build a surrogate for the following function:
@@ -20,11 +16,11 @@ function discont_1D(x)
 end
 ```
 
-Let's choose the MOE Surrogate for 1D. Note that we have to import the `SurrogatesMOE` package in addition to `Surrogates` and `Plots`.
+Let's choose the MOE Surrogate for 1D. Note that we have to import the `GaussianMixtures` in addition to `Surrogates` and `Plots`.
 
 ```@example MOE_1D
 using Surrogates
-using SurrogatesMOE
+using GaussianMixtures
 using Plots
 
 lb = -1.0
@@ -66,7 +62,7 @@ First, we create Gaussian Mixture Models for the number of expert types provided
 
 ```@example MOE_ND
 using Surrogates
-using SurrogatesMOE
+using GaussianMixtures
 
 # helper to test accuracy of predictors
 function rmse(a, b)

diff --git a/docs/src/neural.md b/docs/src/neural.md
@@ -1,9 +1,5 @@
 # Neural Network Surrogate Tutorial
 
-!!! note
-
-    This surrogate requires the 'SurrogatesFlux' module, which can be added by inputting "]add SurrogatesFlux" from the Julia command line.
-
 It's possible to define a neural network as a surrogate, using Flux.
 This is useful because we can call optimization methods on it.
 
@@ -14,7 +10,6 @@ using Plots
 default(c = :matter, legend = false, xlabel = "x", ylabel = "y")
 using Surrogates
 using Flux
-using SurrogatesFlux
 
 function schaffer(x)
     x1 = x[1]

diff --git a/docs/src/polychaos.md b/docs/src/polychaos.md
@@ -1,9 +1,5 @@
 # Polynomial Chaos Surrogate Tutorial
 
-!!! note
-
-    This surrogate requires the 'SurrogatesPolyChaos' module which can be added by inputting "]add SurrogatesPolyChaos" from the Julia command line.
-
 We can create a surrogate using a polynomial expansion, with a different polynomial basis depending on the distribution of the data we are trying to fit. Under the hood, PolyChaos.jl has been used. It is possible to specify a type of polynomial for each dimension of the problem.
 
 ## Sampling
@@ -12,7 +8,7 @@ We choose to sample f in 100 points between 0 and 10 using the `sample` function
 
 ```@example polychaos
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 
 n = 100
@@ -30,7 +26,7 @@ plot!(f, label = "True function", xlims = (lower_bound, upper_bound), legend = :
 ```@example polychaos
 poly1 = PolynomialChaosSurrogate(x, y, lower_bound, upper_bound)
 poly2 = PolynomialChaosSurrogate(
-    x, y, lower_bound, upper_bound, orthopolys = SurrogatesPolyChaos.GaussOrthoPoly(5))
+    x, y, lower_bound, upper_bound, orthopolys = GaussOrthoPoly(5))
 plot(x, y, seriestype = :scatter, label = "Sampled points",
     xlims = (lower_bound, upper_bound), legend = :top)
 plot!(f, label = "True function", xlims = (lower_bound, upper_bound), legend = :top)

diff --git a/docs/src/rosenbrock.md b/docs/src/rosenbrock.md
@@ -9,7 +9,7 @@ Let's import Surrogates and Plots:
 
 ```@example rosen
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 ```
 

diff --git a/docs/src/surrogate.md b/docs/src/surrogate.md
@@ -32,19 +32,19 @@ LobachevskySurrogate(x,y,lb,ub; alpha = collect(one.(x[1])),n::Int = 4, sparse =
 lobachevsky_integral(loba::LobachevskySurrogate,lb,ub)
 ```
 
-  - Support vector machine surrogate, requires `using LIBSVM` and `using SurrogatesSVM`
+  - Support vector machine surrogate, requires `using LIBSVM`.
 
 ```
 SVMSurrogate(x,y,lb::Number,ub::Number)
 ```
 
-  - Random forest surrogate, requires `using XGBoost` and `using SurrogatesRandomForest`
+  - Random forest surrogate, requires `using XGBoost`.
 
 ```
-RandomForestSurrogate(x,y,lb,ub;num_round::Int = 1)
+XGBoostSurrogate(x,y,lb,ub;num_round::Int = 1)
 ```
 
-  - Neural network surrogate, requires `using Flux` and `using SurrogatesFlux`
+  - Neural network surrogate, requires `using Flux`.
 
 ```
 NeuralSurrogate(x,y,lb,ub; model = Chain(Dense(length(x[1]),1), first), loss = (x,y) -> Flux.mse(model(x), y),opt = Descent(0.01),n_echos::Int = 1)

diff --git a/docs/src/tutorials.md b/docs/src/tutorials.md
@@ -123,7 +123,6 @@ Basic example of fitting a neural network on a simple function of two variables.
 using Surrogates
 using Flux
 using Statistics
-using SurrogatesFlux
 
 f = x -> x[1]^2 + x[2]^2
 # Flux models are in single precision by default.

diff --git a/docs/src/water_flow.md b/docs/src/water_flow.md
@@ -7,7 +7,7 @@ It has 8 dimensions.
 
 ```@example water
 using Surrogates
-using SurrogatesPolyChaos
+using PolyChaos
 using Plots
 using LinearAlgebra
 ```