.

Author: tlienart

src/glr/constructors.jl

@@ -139,7 +139,7 @@ With `scale_penalty_with_samples = false` the objective function is
 ``L(y, Xθ) + λ|θ|₂²/2 + γ|θ|₁``.
 """
 function LogisticRegression(
-            λ::Real=1.0,
+            λ::Real=eps(),
             γ::Real=0.0;
             # kwargs
             lambda::Real=λ,

test/fit/quantile.jl

@@ -92,7 +92,7 @@ y1a  = outlify(y1, 0.1)
         θ_ista     = fit(rr, X, y1a, solver=ISTA())
         θ_qr_lasso = rcopy(QUANTREG.rq_fit_lasso(X1, y1a))[:coefficients]
         @test isapprox(J(θ_ls),       888.3748, rtol=1e-5)
-        @test isapprox(J(θ_qr_lasso), 425.5264, rtol=1e-5)
+        @test isapprox(J(θ_qr_lasso), 425.5,    rtol=1e-3)
         # Our algorithms are close enough
         @test isapprox(J(θ_fista),    425.0526, rtol=1e-5)
         @test isapprox(J(θ_ista),     425.4113, rtol=1e-5)

test/runtests.jl

@@ -2,7 +2,7 @@ using MLJLinearModels, Test, LinearAlgebra
 using Random, StableRNGs, DataFrames, ForwardDiff
 import MLJBase # not MLJModelInterface, to mimic the full interface
 
-DO_COMPARISONS = false; include("testutils.jl")
+DO_COMPARISONS = true; include("testutils.jl")
 
 m("UTILS"); include("utils.jl")