shoot me aljeblieft

src/generators/gradient_based/probe.jl

@@ -31,43 +31,3 @@ function ProbeGenerator(;
     user_loss = Objectives.losses_catalogue[loss]
     return GradientBasedGenerator(; loss=user_loss, penalty=penalty, λ=λ, kwargs...)
 end
-
-"""
-    invalidation_rate(ce::AbstractCounterfactualExplanation)
-
-Calculate the invalidation rate of a counterfactual explanation.
-
-# Arguments
-- `ce::AbstractCounterfactualExplanation`: The counterfactual explanation to calculate the invalidation rate for.
-- `kwargs`: Additional keyword arguments to pass to the function.
-
-# Returns
-The invalidation rate of the counterfactual explanation.
-"""
-function invalidation_rate(ce::AbstractCounterfactualExplanation)
-    if !hasfield(typeof(ce.convergence), :invalidation_rate)
-        @warn "Invalidation rate is only defined for InvalidationRateConvergence. Returning 0."
-        return 0.0
-    end
-
-    index_target = findfirst(map(x -> x == ce.target, ce.data.y_levels))
-    f_loss = logits(ce.M, CounterfactualExplanations.decode_state(ce))[index_target]
-    grad = []
-    for i in 1:length(ce.s′)
-        push!(
-            grad,
-            Flux.gradient(
-                () -> logits(ce.M, CounterfactualExplanations.decode_state(ce))[i],
-                Flux.params(ce.s′),
-            )[ce.s′],
-        )
-    end
-    gradᵀ = LinearAlgebra.transpose(grad)
-
-    identity_matrix = LinearAlgebra.Matrix{Float32}(I, length(grad), length(grad))
-    denominator = sqrt(gradᵀ * ce.convergence.variance * identity_matrix * grad)[1]
-
-    normalized_gradient = f_loss / denominator
-    ϕ = Distributions.cdf(Distributions.Normal(0, 1), normalized_gradient)
-    return 1 - ϕ
-end

src/objectives/Objectives.jl

@@ -1,7 +1,6 @@
 module Objectives
 
 using ..CounterfactualExplanations
-using ..CounterfactualExplanations.Generators
 using Flux
 using Flux.Losses
 using ChainRulesCore

src/objectives/loss_functions.jl

@@ -45,6 +45,46 @@ function Flux.Losses.mse(ce::AbstractCounterfactualExplanation; kwargs...)
     return loss
 end
 
+"""
+    invalidation_rate(ce::AbstractCounterfactualExplanation)
+
+Calculate the invalidation rate of a counterfactual explanation.
+
+# Arguments
+- `ce::AbstractCounterfactualExplanation`: The counterfactual explanation to calculate the invalidation rate for.
+- `kwargs`: Additional keyword arguments to pass to the function.
+
+# Returns
+The invalidation rate of the counterfactual explanation.
+"""
+function invalidation_rate(ce::AbstractCounterfactualExplanation)
+    if !hasfield(typeof(ce.convergence), :invalidation_rate)
+        @warn "Invalidation rate is only defined for InvalidationRateConvergence. Returning 0."
+        return 0.0
+    end
+
+    index_target = findfirst(map(x -> x == ce.target, ce.data.y_levels))
+    f_loss = logits(ce.M, CounterfactualExplanations.decode_state(ce))[index_target]
+    grad = []
+    for i in 1:length(ce.s′)
+        push!(
+            grad,
+            Flux.gradient(
+                () -> logits(ce.M, CounterfactualExplanations.decode_state(ce))[i],
+                Flux.params(ce.s′),
+            )[ce.s′],
+        )
+    end
+    gradᵀ = LinearAlgebra.transpose(grad)
+
+    identity_matrix = LinearAlgebra.Matrix{Float32}(I, length(grad), length(grad))
+    denominator = sqrt(gradᵀ * ce.convergence.variance * identity_matrix * grad)[1]
+
+    normalized_gradient = f_loss / denominator
+    ϕ = Distributions.cdf(Distributions.Normal(0, 1), normalized_gradient)
+    return 1 - ϕ
+end
+
 """
     hinge_loss_ir(convergence::InvalidationRateConvergence, ce::AbstractCounterfactualExplanation)