progres on granular examination

notebooks/synthetic.qmd

@@ -18,13 +18,14 @@ datasets = Dict(
 )
 
 # Hyperparameters:
-cvgs = [0.5, 0.75, 0.90, 0.95, 0.99]
-temps = [0.05, 0.1, 0.5, 1.0, 2.0]
-Λ = [0.1, 1.0, 5.0]
+cvgs = [0.5, 0.75, 0.95]
+temps = [0.01, 0.1, 1.0]
+Λ = [0.0, 0.1, 1.0]
+l2_λ = 0.1
 
 # Classifiers:
-epochs = 100
-link_fun = sigmoid
+epochs = 250
+link_fun = relu
 logreg = NeuralNetworkClassifier(builder=MLJFlux.Linear(σ=link_fun), epochs=epochs)
 mlp = NeuralNetworkClassifier(builder=MLJFlux.MLP(hidden=(32,), σ=link_fun), epochs=epochs)
 ensmbl = EnsembleModel(model=mlp, n=5)
@@ -37,6 +38,9 @@ classifiers = Dict(
 # Search parameters:
 target = 2
 factual = 1
+max_iter = 50
+gradient_tol = 1e-2
+opt = Descent(0.01)
 ```
 
 ```{julia}
@@ -46,6 +50,7 @@ for (dataname, data) in datasets
     # Data:
     X = table(permutedims(data.X))
     y = data.output_encoder.labels
+    x = select_factual(data,rand(1:size(data.X,2)))
 
     for (clf_name, clf) in classifiers, cov in cvgs
 
@@ -56,21 +61,21 @@ for (dataname, data) in datasets
         M = CCE.ConformalModel(mach.model, mach.fitresult)
 
         # Set up CCE:
-        yhat = predict_label(M, data)
-        factual_label = data.y_levels[factual]
-        target_label = data.y_levels[target]
-        x = select_factual(data,rand(findall(yhat .== factual_label)))
+        factual_label = predict_label(M, data, x)[1]
+        target_label = data.y_levels[data.y_levels .!= factual_label][1]
 
         for λ in Λ, temp in temps
 
             # CCE for given classifier, coverage, temperature and λ:
-            generator = CCEGenerator(temp=temp, λ=λ)
+            generator = CCEGenerator(temp=temp, λ=[l2_λ,λ], opt=opt)
             @assert predict_label(M, data, x) != target_label
             ce = try
                 generate_counterfactual(
                     x, target_label, data, M, generator;
                     initialization=:identity,
                     converge_when=:generator_conditions,
+                    gradient_tol=gradient_tol,
+                    max_iter=max_iter,
                 )
             catch
                 missing
@@ -82,7 +87,7 @@ for (dataname, data) in datasets
                 coverage = cov,
                 temperature = temp,
                 λ = λ,
-                counterfactual = ce,
+                ce = ce,
                 factual = factual_label,
                 target = target_label,
             )
@@ -96,5 +101,30 @@ end
 ```
 
 ```{julia}
+function plot_ce(results; dataset=:multi_class, classifier=:mlp, λ=0.1, img_height=300, zoom=-0.2)
+    df_plot = results[results.dataset .== dataset,:] |>
+        res -> res[res.classifier .== classifier,:] |>
+        res -> res[res.λ .== λ,:]
+    plts = map(eachrow(df_plot)) do row
+        Plots.plot(
+            row.ce,
+            title="cov: $(row.coverage), temp: $(row.temperature)",
+            cbar=false,
+            zoom=zoom,
+            legend=false,
+        )
+    end
+    nrow = length(cvgs)
+    ncol = length(temps)
+    _layout = (nrow, ncol)
+    Plots.plot(
+        plts..., 
+        size=img_height.*reverse(_layout), layout=_layout,
+        plot_title="λ: $λ, dataset: $dataset, classifier: $classifier",
+    )
+end
+```
 
+```{julia}
+plot_ce(results; λ=Λ[1], dataset=:moons)
 ```

src/generator.jl

 using CounterfactualExplanations.Objectives
 
 "Constructor for `CCEGenerator`."
-function CCEGenerator(; λ::Union{AbstractFloat,Vector{<:AbstractFloat}}=[0.1, 1.0], κ::Real=0.0, temp::Real=0.05, kwargs...)
+function CCEGenerator(; λ::Union{AbstractFloat,Vector{<:AbstractFloat}}=[0.1, 1.0], κ::Real=1.0, temp::Real=0.05, kwargs...)
     function _set_size_penalty(ce::AbstractCounterfactualExplanation)
         return CCE.set_size_penalty(ce; κ=κ, temp=temp)
     end