Pass keywords to glmnet (#19)

* actually pass keywords to glmnet

* test passing of keywords to glmnet

src/lasso.jl

@@ -1,11 +1,11 @@
 function fit_lasso_path(
     mm, presences;
-    wts, penalty_factor, λ, kw...) 
+    weights, penalty_factor, lambda, kw...) 
 
     presence_matrix = [1 .- presences presences]
     GLMNet.glmnet(
         mm, presence_matrix, GLMNet.Binomial(); 
-        weights = wts, penalty_factor = penalty_factor, lambda = λ, standardize = false)
+        weights, penalty_factor, lambda, standardize = false, kw...)
 end
 
 get_coefs(path::GLMNet.GLMNetPath) = path.betas

src/maxnet_function.jl

@@ -61,7 +61,6 @@ function maxnet(
         kw...
     )
 end
-#maxnet(presences, predictors; kw...) = maxnet(presences, predictors, features; kw...)
 
 ### internal methods where features is not a keyword
 
@@ -129,10 +128,10 @@ function _maxnet(
     weights = presences .* 1. .+ (1 .- presences) .* weight_factor
 
     # generate lambdas
-    λ = lambdas(reg, presences, weights; λmax = 4, n = 200)
+    lambda = lambdas(reg, presences, weights; λmax = 4, n = 200)
 
     # Fit the model
-    lassopath = fit_lasso_path(mm, presences, wts = weights, penalty_factor = reg, λ = λ)
+    lassopath = fit_lasso_path(mm, presences; weights, penalty_factor = reg, lambda, kw...)
 
     # get the coefficients out
     coefs = SparseArrays.sparse(get_coefs(lassopath)[:, end])

test/runtests.jl

@@ -31,6 +31,7 @@ env1 = map(e -> [e[1]], env) # just the first row
     @test predictors == (a = [1,2,3,1], b = [1,2,3,1])
 end
 
+
 @testset "Maxnet" begin
     # some class combinations and keywords
     m = maxnet(p_a, env; features = "lq");
@@ -72,7 +73,14 @@ end
     @test complexity(empty_model) == 0
     @test Maxnet.selected_features(empty_model) == Symbol[]
     @test length(unique(predict(empty_model, env))) == 1
+
+    # test that keywords arguments are passed to glmnet
+    weights = ifelse.(p_a, 1.0, 10.0)
+    m_w = maxnet(p_a, env; features = "lq", addsamplestobackground = false, weights)
+    m = maxnet(p_a, env; features = "lq", addsamplestobackground = false)
+    @test m_w.entropy > m.entropy
 end
+m = maxnet(p_a, env; features = "lq", addsamplestobackground = false)
 
 @testset "MLJ" begin
     using MLJBase