Merge pull request #15 from JuliaDecisionFocusedLearning/update-datasample

Rename DataSample fields

Author: BatyLeo

docs/src/tutorials/warcraft.jl

@@ -25,10 +25,10 @@ train_dataset, test_dataset = dataset[1:45], dataset[46:50]
 sample = test_dataset[1]
 # `x` correspond to the input features, i.e. the input image (3D array) in the Warcraft benchmark case:
 x = sample.x
-# `θ` correspond to the true unknown terrain weights. We use the opposite of the true weights in order to formulate the optimization problem as a maximization problem:
-θ_true = sample.θ
-# `y` correspond to the optimal shortest path, encoded as a binary matrix:
-y_true = sample.y
+# `θ_true` correspond to the true unknown terrain weights. We use the opposite of the true weights in order to formulate the optimization problem as a maximization problem:
+θ_true = sample.θ_true
+# `y_true` correspond to the optimal shortest path, encoded as a binary matrix:
+y_true = sample.y_true
 # `instance` is not used in this benchmark, therefore set to nothing:
 isnothing(sample.instance)
 
@@ -50,7 +50,7 @@ maximizer = generate_maximizer(b; dijkstra=true)
 # In the case o fthe Warcraft benchmark, the method has an additional keyword argument to chose the algorithm to use: Dijkstra's algorithm or Bellman-Ford algorithm.
 y = maximizer(θ)
 # As we can see, currently the pipeline predicts random noise as cell weights, and therefore the maximizer returns a straight line path.
-plot_data(b, DataSample(; x, θ, y))
+plot_data(b, DataSample(; x, θ_true=θ, y_true=y))
 # We can evaluate the current pipeline performance using the optimality gap metric:
 starting_gap = compute_gap(b, test_dataset, model, maximizer)
 
@@ -70,7 +70,7 @@ opt_state = Flux.setup(Adam(1e-3), model)
 loss_history = Float64[]
 for epoch in 1:50
     val, grads = Flux.withgradient(model) do m
-        sum(loss(m(sample.x), sample.y) for sample in train_dataset) / length(train_dataset)
+        sum(loss(m(x), y_true) for (; x, y_true) in train_dataset) / length(train_dataset)
     end
     Flux.update!(opt_state, model, grads[1])
     push!(loss_history, val)
@@ -85,4 +85,4 @@ final_gap = compute_gap(b, test_dataset, model, maximizer)
 #
 θ = model(x)
 y = maximizer(θ)
-plot_data(b, DataSample(; x, θ, y))
+plot_data(b, DataSample(; x, θ_true=θ, y_true=y))

src/FixedSizeShortestPath/FixedSizeShortestPath.jl

@@ -132,7 +132,10 @@ function Utils.generate_dataset(
 
     # Label solutions
     solutions = shortest_path_maximizer.(costs)
-    return [DataSample(; x=x, θ=θ, y=y) for (x, θ, y) in zip(features, costs, solutions)]
+    return [
+        DataSample(; x, θ_true, y_true) for
+        (x, θ_true, y_true) in zip(features, costs, solutions)
+    ]
 end
 
 """

src/PortfolioOptimization/PortfolioOptimization.jl

@@ -110,7 +110,10 @@ function Utils.generate_dataset(
     maximizer = Utils.generate_maximizer(bench)
     solutions = maximizer.(costs)
 
-    return [DataSample(; x=x, θ=θ, y=y) for (x, θ, y) in zip(features, costs, solutions)]
+    return [
+        DataSample(; x, θ_true, y_true) for
+        (x, θ_true, y_true) in zip(features, costs, solutions)
+    ]
 end
 
 """

src/SubsetSelection/SubsetSelection.jl

@@ -73,7 +73,10 @@ function Utils.generate_dataset(
         mapping.(features)
     end
     solutions = top_k.(costs, k)
-    return [DataSample(; x=x, θ=θ, y=y) for (x, θ, y) in zip(features, costs, solutions)]
+    return [
+        DataSample(; x, θ_true, y_true) for
+        (x, θ_true, y_true) in zip(features, costs, solutions)
+    ]
 end
 
 """

src/Utils/data_sample.jl

@@ -9,10 +9,10 @@ $TYPEDFIELDS
 @kwdef struct DataSample{F,S,C,I}
     "features"
     x::F
-    "costs"
-    θ::C = nothing
-    "solution"
-    y::S = nothing
-    "instance"
+    "target cost parameters (optional)"
+    θ_true::C = nothing
+    "target solution (optional)"
+    y_true::S = nothing
+    "instance object (optional)"
     instance::I = nothing
 end

src/Utils/interface.jl

@@ -73,8 +73,8 @@ function compute_gap(
 
     for sample in dataset
         x = sample.x
-        θ̄ = sample.θ
-        ȳ = sample.y
+        θ̄ = sample.θ_true
+        ȳ = sample.y_true
         θ = statistical_model(x)
         y = maximizer(θ)
         target_obj = objective_value(bench, θ̄, ȳ)

src/Warcraft/Warcraft.jl

@@ -87,12 +87,14 @@ The keyword argument `θ_true` is used to set the color range of the weights plo
 function Utils.plot_data(
     ::WarcraftBenchmark,
     sample::DataSample;
-    θ_true=sample.θ,
+    θ_true=sample.θ_true,
     θ_title="Weights",
     y_title="Path",
     kwargs...,
 )
-    (; x, y, θ) = sample
+    x = sample.x
+    y = sample.y_true
+    θ = sample.θ_true
     im = dropdims(x; dims=4)
     img = convert_image_for_plot(im)
     p1 = Plots.plot(

src/Warcraft/utils.jl

@@ -40,7 +40,7 @@ function create_dataset(decompressed_path::String, nb_samples::Int)
     ]
     Y = [BitMatrix(terrain_labels[:, :, i]) for i in 1:N]
     WG = [-terrain_weights[:, :, i] for i in 1:N]
-    return [DataSample(; x, y, θ) for (x, y, θ) in zip(X, Y, WG)]
+    return [DataSample(; x, y_true, θ_true) for (x, y_true, θ_true) in zip(X, Y, WG)]
 end
 
 """

test/fixed_size_shortest_path.jl

@@ -18,9 +18,7 @@
     @test gap >= 0
 
     for sample in dataset
-        x = sample.x
-        θ_true = sample.θ
-        y_true = sample.y
+        (; x, θ_true, y_true) = sample
         @test all(θ_true .< 0)
         @test size(x) == (p,)
         @test length(θ_true) == A

test/portfolio_optimization.jl

@@ -10,9 +10,7 @@
     maximizer = generate_maximizer(b)
 
     for sample in dataset
-        x = sample.x
-        θ_true = sample.θ
-        y_true = sample.y
+        (; x, θ_true, y_true) = sample
         @test size(x) == (p,)
         @test length(θ_true) == d
         @test length(y_true) == d
@@ -24,6 +22,6 @@
 
         y = maximizer(θ)
         @test length(y) == d
-        @test sum(y) <= 1
+        @test sum(y) <= 1 + 1e-6
     end
 end