Merge pull request #507 from JuliaTrustworthyAI/issue-with-benchmark

Fix small issue in benchmark function

CHANGELOG.md

@@ -11,6 +11,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 ### Changed
 
 - Slight change to `FlattenedCE` and `unflatten` to ensure that basic functionality remains intact. [#505]
+- Fixed small issue in `benchmark` function.
 
 ## Version [1.4.1] - 2024-12-19
 

src/evaluation/benchmark.jl

@@ -176,7 +176,8 @@ function benchmark(
     )
 
     # Unflatten for evaluation:
-    @assert typeof(ces) == Vector{FlattenedCE} "Expecting a vector of `FlattenedCE`. Did you accidentally set `return_flattened=false`?"
+    @assert all(typeof.(ces) .== FlattenedCE) "Expecting a vector of `FlattenedCE`. Did you accidentally set `return_flattened=false`?"
+    ces = convert(Vector{FlattenedCE}, ces)
     ces = unflatten_for_eval(ces, data, Ms, gens, kwrgs)
 
     # Meta Data:
@@ -432,7 +433,8 @@ function benchmark(
             )
 
             # Unflatten for evaluation:
-            @assert typeof(ces) == Vector{FlattenedCE} "Expecting a vector of `FlattenedCE`. Did you accidentally set `return_flattened=false`?"
+            @assert all(typeof.(ces) .== FlattenedCE) "Expecting a vector of `FlattenedCE`. Did you accidentally set `return_flattened=false`?"
+            ces = convert(Vector{FlattenedCE}, ces)
             ces = unflatten_for_eval(ces, data, Ms, gens, kwrgs)
 
             # Free up memory:

test/Project.toml

@@ -27,6 +27,7 @@ Serialization = "9e88b42a-f829-5b0c-bbe9-9e923198166b"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
 TaijaData = "9d524318-b4e6-4a65-86d2-b2b72d07866c"
+TaijaParallel = "bf1c2c22-5e42-4e78-8b6b-92e6c673eeb0"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
 
 [compat]

test/other/evaluation.jl

@@ -3,6 +3,7 @@ using CounterfactualExplanations.Evaluation:
 using CounterfactualExplanations.Objectives: distance
 using Serialization: serialize
 using TaijaData: load_moons, load_circles
+using TaijaParallel: ThreadsParallelizer
 
 # Dataset
 data = TaijaData.load_overlapping()
@@ -70,6 +71,17 @@ end
 @testset "Benchmarking" begin
     bmk = Evaluation.benchmark(counterfactual_data; convergence=:generator_conditions)
 
+    @testset "Parallelization" begin
+        @testset "Threads" begin
+            parallelizer = ThreadsParallelizer()
+            bmk = benchmark(
+                counterfactual_data;
+                convergence=:generator_conditions,
+                parallelizer=parallelizer,
+            )
+        end
+    end
+
     @testset "Basics" begin
         @test typeof(bmk()) <: DataFrame
         @test typeof(bmk(; agg=nothing)) <: DataFrame