add tests for params, fix parameter order

Author: itsdfish

Project.toml

@@ -1,7 +1,7 @@
 name = "SequentialSamplingModels"
 uuid = "0e71a2a6-2b30-4447-8742-d083a85e82d1"
 authors = ["itsdfish"]
-version = "0.12.7"
+version = "0.13.0"
 
 [deps]
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"

src/multi_choice_models/AttentionalDiffusion.jl

@@ -83,6 +83,8 @@ function aDDM(; ν = [5.0, 4.0], α = 1.0, z = 0.0, θ = 0.3, σ = 0.02, Δ = 0.
     return aDDM(ν, σ, Δ, θ, α, z, τ)
 end
 
+params(d::aDDM) = (d.ν, d.σ, d.Δ, d.θ, d.α, d.z, d.τ)
+
 get_pdf_type(d::aDDM) = Approximate
 
 """

src/multi_choice_models/RDM.jl

@@ -121,8 +121,8 @@ struct RDM{T <: Real} <: AbstractRDM
     τ::T
 end
 
-function RDM(ν, k::T, A, τ) where {T}
-    _, A, k, τ = promote(ν[1], k, A, τ)
+function RDM(ν, A, k::T, τ) where {T}
+    _, A, k, τ = promote(ν[1], A, k, τ)
     ν = convert(Vector{T}, ν)
     return RDM(ν, A, k, τ)
 end
@@ -131,7 +131,7 @@ function params(d::AbstractRDM)
     (d.ν, d.A, d.k, d.τ)
 end
 
-RDM(; ν = [1, 2], k = 0.3, A = 0.7, τ = 0.2) = RDM(ν, k, A, τ)
+RDM(; ν = [1, 2], k = 0.3, A = 0.7, τ = 0.2) = RDM(ν, A, k, τ)
 
 function rand(rng::AbstractRNG, dist::AbstractRDM)
     (; ν, A, k, τ) = dist

src/multi_choice_models/maaDDM.jl

@@ -125,6 +125,8 @@ function maaDDM(;
     return maaDDM(ν, σ, Δ, θ, ϕ, ω, α, z, τ)
 end
 
+params(d::maaDDM) = (d.ν, d.σ, d.Δ, d.θ, d.ϕ, d.ω, d.α, d.z, d.τ)
+
 get_pdf_type(d::maaDDM) = Approximate
 
 n_options(d::maaDDM) = size(d.ν, 1)

src/multi_choice_models/stDDM.jl

@@ -7,18 +7,27 @@ An object for the starting-time diffusion decision model.
 
 - `ν::T`:  vector of drift rate weights for attribute one and two. ν ∈ ℝ.
 - `σ::T`:  diffusion noise. σ ∈ ℝ⁺.
-- `s::T`:  initial latency bias (positive for attribute two, negative for attribute one)
-- `z::T`:  initial evidence. z ∈ [0,α].  
 - `η::T`:  vector of variability in drift rate for attribute one and two. η ∈ ℝ⁺.
+- `s::T`:  initial latency bias (positive for attribute two, negative for attribute one)
 - `ρ::T`:  correlation between drift rate for attributes. ρ ∈ [-1,1].
 - `α::T`:  evidence threshold. α ∈ ℝ⁺.
+- `z::T`:  initial evidence. z ∈ [0,α].  
 - `τ::T`:  non-decision time. τ ∈ [0, min_rt].
 
 # Constructors 
 
-    stDDM(ν, σ, s, z, η, ρ, α, τ)
-
-    stDDM(;ν = [0.5,0.6],σ = 1,s = 0.50, z = 0.50, η = [1.0,1.0], ρ = 0.00, α = 1.0, τ = .300)
+    stDDM(ν, σ, η, s, ρ, α, z, τ)
+
+    stDDM(;
+        ν = [0.5, 0.6],
+        σ = 1,
+        s = 0.50,
+        ρ = 0.0,
+        η = fill(1.0, length(ν)),
+        α = 1.0,
+        z = 0.50,
+        τ = 0.300
+    )
         
 # Example 
 
@@ -27,15 +36,15 @@ using SequentialSamplingModels
 
 ν = [0.5, 0.6]
 σ = 1
-s = 0.50
-z = 0.50
 η = [1.0, 1.0]
 ρ = 0.00
+s = 0.50
 α = 1.0
+z = 0.50
 τ = 0.300
 
 # Create stDDM model instance
-dist = stDDM(;ν, σ, s, z, η, ρ, α, τ)
+dist = stDDM(; ν, σ, s, z, η, ρ, α, τ)
 
 choices,rts = rand(dist, 500)
 ```
@@ -56,36 +65,36 @@ Sullivan, N.J., Huettel, S.A. Healthful choices depend on the latency and rate o
 mutable struct stDDM{T <: Real} <: AbstractstDDM
     ν::Vector{T}
     σ::T
-    s::T
-    z::T
     η::Vector{T}
+    s::T
     ρ::T
     α::T
+    z::T
     τ::T
 end
 
-function stDDM(ν, σ, s, z, η, ρ, α, τ::T) where {T}
-    _, σ, s, z, _, ρ, α, τ = promote(ν[1], σ, s, z, η[1], ρ, α, τ)
+function stDDM(ν, σ, η, s, ρ, α, z, τ::T) where {T}
+    _, σ, _, s, ρ, α, z, τ = promote(ν[1], σ, η[1], s, ρ, α, z, τ)
     ν = convert(Vector{T}, ν)
     η = convert(Vector{T}, η)
-    return stDDM(ν, σ, s, z, η, ρ, α, τ)
+    return stDDM(ν, σ, η, s, ρ, α, z, τ)
 end
 
 function stDDM(;
     ν = [0.5, 0.6],
     σ = 1,
     s = 0.50,
-    z = 0.50,
-    η = fill(1.0, length(ν)),
     ρ = 0.0,
+    η = fill(1.0, length(ν)),
     α = 1.0,
+    z = 0.50,
     τ = 0.300
 )
-    return stDDM(ν, σ, s, z, η, ρ, α, τ)
+    return stDDM(ν, σ, η, s, ρ, α, z, τ)
 end
 
 function params(d::AbstractstDDM)
-    (d.ν, d.σ, d.s, d.z, d.η, d.ρ, d.α, d.τ)
+    (d.ν, d.σ, d.η, d.s, d.ρ, d.α, d.z, d.τ)
 end
 
 get_pdf_type(d::AbstractstDDM) = Approximate

test/alternative_geometries/circular_ddm.jl

@@ -286,4 +286,14 @@
         _, max_idx = findmax(LLs)
         @test parms.η[2] ≈ η2s[max_idx] atol = 0.05
     end
+
+    @safetestset "params" begin
+        using Test
+        using Distributions
+        using SequentialSamplingModels
+
+        parms = (; ν = [1, 0.5], σ = 1, η = [1, 1], α = 1.5, τ = 0.30)
+        model = CDDM(; parms...)
+        @test values(parms) == params(model)
+    end
 end

test/multi_choice_models/attentional_diffusion_tests.jl

@@ -93,4 +93,14 @@
 
         @test evidence[end] ≈ -1 atol = 0.030
     end
+
+    @safetestset "params" begin
+        using Test
+        using Distributions
+        using SequentialSamplingModels
+
+        parms = (; ν = [5.0, 4.0], σ = 0.02, Δ = 0.0004, θ = 0.3, α = 1.0, z = 0.0, τ = 0.0)
+        model = aDDM(; parms...)
+        @test values(parms) == params(model)
+    end
 end

test/multi_choice_models/ddm_tests.jl

@@ -311,4 +311,14 @@
         time_steps, evidence = simulate(rng, dist; Δt = 0.0001)
         @test evidence[end] ≈ 0.0 atol = 0.010
     end
+
+    @safetestset "params" begin
+        using Test
+        using Distributions
+        using SequentialSamplingModels
+
+        parms = (; ν = 1.00, α = 0.80, z = 0.50, τ = 0.30)
+        model = DDM(; parms...)
+        @test values(parms) == params(model)
+    end
 end

test/multi_choice_models/lba_tests.jl

@@ -145,4 +145,14 @@
             end
         end
     end
+
+    @safetestset "params" begin
+        using Test
+        using Distributions
+        using SequentialSamplingModels
+
+        parms = (; ν = [2.0, 1.75], σ = 1, A = 0.8, k = 0.5, τ = 0.3)
+        model = LBA(; parms...)
+        @test values(parms) == params(model)
+    end
 end

test/multi_choice_models/lca_tests.jl

@@ -142,4 +142,14 @@
         @test size(evidence, 2) == 2
         @test maximum(evidence[end, :]) ≈ α atol = 0.005
     end
+
+    @safetestset "params" begin
+        using Test
+        using Distributions
+        using SequentialSamplingModels
+
+        parms = (; ν = [2.5, 2.0], σ = 1.0, β = 0.20, λ = 0.10, α = 1.5, τ = 0.30)
+        model = LCA(; parms...)
+        @test values(parms) == params(model)
+    end
 end