drop models as deps and add compats

.github/workflows/CI.yml

@@ -19,7 +19,8 @@ jobs:
       fail-fast: false
       matrix:
         version:
-          - '1.9'
+          - '1.10'
+          - '1'
         os:
           - ubuntu-latest
         arch:

Project.toml

@@ -5,27 +5,19 @@ version = "1.0.0-DEV"
 
 [deps]
 CategoricalArrays = "324d7699-5711-5eae-9e2f-1d82baa6b597"
-CategoricalDistributions = "af321ab8-2d2e-40a6-b165-3d674595d28e"
 ComputationalResources = "ed09eef8-17a6-5b46-8889-db040fac31e3"
-DecisionTree = "7806a523-6efd-50cb-b5f6-3fa6f1930dbb"
-DimensionalData = "0703355e-b756-11e9-17c0-8b28908087d0"
 Distances = "b4f34e82-e78d-54a5-968a-f98e89d6e8f7"
-EvoTrees = "f6006082-12f8-11e9-0c9c-0d5d367ab1e5"
 GLM = "38e38edf-8417-5370-95a0-9cbb8c7f171a"
 GeoInterface = "cf35fbd7-0cd7-5166-be24-54bfbe79505f"
-Lasso = "b4fcebef-c861-5a0f-a7e2-ba9dc32b180a"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 Loess = "4345ca2d-374a-55d4-8d30-97f9976e7612"
 MLJBase = "a7f614a8-145f-11e9-1d2a-a57a1082229d"
-MLJDecisionTreeInterface = "c6f25543-311c-4c74-83dc-3ea6d1015661"
-MLJGLMInterface = "caf8df21-4939-456d-ac9c-5fefbfb04c0c"
 PrettyTables = "08abe8d2-0d0c-5749-adfa-8a2ac140af0d"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 Rasters = "a3a2b9e3-a471-40c9-b274-f788e487c689"
 ScientificTypesBase = "30f210dd-8aff-4c5f-94ba-8e64358c1161"
 Shapley = "855ca7ad-a6ef-4de2-9ca8-726fe2a39065"
 StatisticalMeasures = "a19d573c-0a75-4610-95b3-7071388c7541"
-StatisticalMeasuresBase = "c062fc1d-0d66-479b-b6ac-8b44719de4cc"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 StatsAPI = "82ae8749-77ed-4fe6-ae5f-f523153014b0"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
@@ -40,20 +32,37 @@ Makie = "ee78f7c6-11fb-53f2-987a-cfe4a2b5a57a"
 SpeciesDistributionModelsMakieExt = "Makie"
 
 [compat]
-CategoricalDistributions = "0.1.14"
+CategoricalArrays = "0.10"
+ComputationalResources = "0.3"
 Distances = "0.10"
-MLJGLMInterface = "0.3.7"
-Rasters = "0.10.1, 0.11, 0.12"
+Distributions = "0.25"
+GLM = "1.9.0"
+GeoInterface = "1"
+Loess = "0.6"
+MLJBase = "1.7.0"
+Makie = "0.20, 0.21"
+PrettyTables = "2"
+Rasters = "0.12"
+ScientificTypesBase = "3"
+Shapley = "0.1"
 StatisticalMeasures = "0.1.5"
+StatsAPI = "1.7.0"
+StatsBase = "0.34"
 StatsModels = "0.7.3"
-julia = "1.9"
+Tables = "1"
+Test = "1"
+ThreadsX = "0.1.12"
+julia = "1.10"
 
 [extras]
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
+EvoTrees = "f6006082-12f8-11e9-0c9c-0d5d367ab1e5"
+MLJDecisionTreeInterface = "c6f25543-311c-4c74-83dc-3ea6d1015661"
+MLJGLMInterface = "caf8df21-4939-456d-ac9c-5fefbfb04c0c"
 MLJModels = "d491faf4-2d78-11e9-2867-c94bc002c0b7"
 Makie = "ee78f7c6-11fb-53f2-987a-cfe4a2b5a57a"
 StableRNGs = "860ef19b-820b-49d6-a774-d7a799459cd3"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [targets]
-test = ["Distributions", "Makie", "MLJModels", "StableRNGs", "Test"]
+test = ["Distributions", "EvoTrees", "Makie", "MLJDecisionTreeInterface", "MLJGLMInterface", "MLJModels", "StableRNGs", "Test"]

src/SpeciesDistributionModels.jl

@@ -1,18 +1,15 @@
 module SpeciesDistributionModels
 
 import Tables, StatsBase, Statistics, StatsAPI, StatsModels, LinearAlgebra, Random, ThreadsX
-import MLJBase, StatisticalMeasures, StatisticalMeasuresBase, ScientificTypesBase, CategoricalArrays
-import GLM, PrettyTables, Rasters, EvoTrees, DecisionTree, Shapley, Loess, Distances
+import MLJBase, StatisticalMeasures, CategoricalArrays
+import GLM, PrettyTables, Rasters, Shapley, Loess, Distances
 import GeoInterface as GI
 
-using MLJBase: pdf
-using Rasters: Raster, RasterStack, Band
-using ScientificTypesBase: Continuous, OrderedFactor, Multiclass, Count
+using Rasters: Raster, RasterStack, Band, DD
 using ComputationalResources: CPU1, CPUThreads, AbstractCPU, CPUProcesses
-
+using StatisticalMeasures: auc, kappa, sensitivity, selectivity, accuracy, StatisticalMeasuresBase
 using ScientificTypesBase: Continuous, OrderedFactor, Multiclass, Count
-using StatisticalMeasures: auc, kappa, sensitivity, selectivity, accuracy
-import MLJBase: StratifiedCV, CV, Holdout, ResamplingStrategy, Machine, Probabilistic
+import MLJBase: StratifiedCV, CV, Holdout, ResamplingStrategy, Machine, Probabilistic, pdf
 
 export SDMensemble, predict, sdm, sdmdata, select, machines, machine_keys,
     remove_collinear, thin,
@@ -25,14 +22,12 @@ export auc, kappa, sensitivity, selectivity, accuracy,
     Continuous, OrderedFactor, Multiclass, Count,
     StratifiedCV, CV, Holdout, ResamplingStrategy
 #include("learningnetwork.jl")
-include("models.jl")
 include("data_utils.jl")
 include("resample.jl")
 # export stubs for extensions
 export interactive_response_curves, interactive_evaluation
 
 include("collinearity.jl")
-include("models.jl")
 include("ensemble.jl")
 include("predict.jl")
 include("explain/explain.jl")

test/runtests.jl

@@ -3,6 +3,10 @@ import SpeciesDistributionModels as SDM
 using StableRNGs, Distributions, Test
 using Makie
 
+using MLJGLMInterface: LinearBinaryClassifier
+using EvoTrees: EvoTreeClassifier
+using MLJDecisionTreeInterface: RandomForestClassifier
+
 rng = StableRNG(0)
 #using Random; rng = Random.GLOBAL_RNG
 # some mock data
@@ -22,10 +26,10 @@ presencedata = (a = rand(rng, n), b = rand(rng, n).^2, c = sqrt.(rand(rng, n)))
 
     ## ensemble
     models = (
-        rf = SDM.random_forest(; rng),
-        rf2 = OneHotEncoder() |> SDM.random_forest(; max_depth = 3, rng),
-        lm = SDM.linear_model(),
-        brt = SDM.boosted_regression_tree(; rng)
+        rf = RandomForestClassifier(; rng),
+        rf2 = OneHotEncoder() |> RandomForestClassifier(; max_depth = 3, rng),
+        lm = LinearBinaryClassifier(),
+        brt = EvoTreeClassifier(; rng)
     )
 
     ensemble = sdm(data, models;