add formatter

Author: itsdfish

.JuliaFormatter.toml

@@ -0,0 +1,29 @@
+always_use_return = false
+annotate_untyped_fields_with_any = true
+conditional_to_if = false
+import_to_using = false
+join_lines_based_on_source = true
+normalize_line_endings = "unix"
+pipe_to_function_call = false
+remove_extra_newlines = true
+short_to_long_function_def = false
+long_to_short_function_def = false
+whitespace_in_kwargs = true
+whitespace_ops_in_indices = true
+whitespace_typedefs = true
+indent = 4
+margin = 92
+format_docstrings = false
+align_struct_field = false
+align_assignment = false
+align_conditional = false
+align_pair_arrow = false
+align_matrix = false
+trailing_comma = false
+trailing_zero = true
+indent_submodule = false
+separate_kwargs_with_semicolon = false
+surround_whereop_typeparameters = true
+variable_call_indent = []
+yas_style_nesting = false
+disallow_single_arg_nesting = true

sandbox/sandbox.jl

@@ -2,25 +2,25 @@
 #                                         load packages
 ############################################################################################################
 cd(@__DIR__)
-using Pkg 
+using Pkg
 Pkg.activate("..")
 using CategorizationModels
-using Random 
-using Test 
+using Random
+using Test
 
 Random.seed!(5)
 
 # model predictions 
 parms = (μ = 80.0,
-        σ = 20.0,
-        υ_k_k = 1.0,
-        υ_s_k = 2.0,
-        υ_k_s = 3.0,
-        υ_s_s = 4.0,
-        λ_k_k = .2,
-        λ_s_k = .3,
-        λ_k_s = .4,
-        λ_s_s = .5)
+    σ = 20.0,
+    υ_k_k = 1.0,
+    υ_s_k = 2.0,
+    υ_k_s = 3.0,
+    υ_s_s = 4.0,
+    λ_k_k = 0.2,
+    λ_s_k = 0.3,
+    λ_k_s = 0.4,
+    λ_s_s = 0.5)
 
 # number of evidence states 
 n_states = 96
@@ -29,7 +29,7 @@ n_states = 96
 n_options = 6
 
 # create a model object 
-model = QuantumModel(;parms..., n_states)
+model = QuantumModel(; parms..., n_states)
 
 # generate predictions for all conditions 
 preds = predict(model, n_options)
@@ -40,17 +40,15 @@ data = rand(model, preds, 100)
 # compute the logpdf of each data point
 LLs = logpdf(model, preds, data)
 
-
-
 function make_hamiltonian1(n_states, υ, σ)
     mat = zeros(n_states, n_states)
     for c ∈ 1:n_states, r ∈ 1:n_states
         if (c - r) == 1
-            mat[r,c] = σ
+            mat[r, c] = σ
         elseif (c - r) == -1
-            mat[r,c] = σ
+            mat[r, c] = σ
         elseif r == c
-            mat[r,c] = (r / n_states) * υ
+            mat[r, c] = (r / n_states) * υ
         end
     end
     return mat
@@ -59,10 +57,10 @@ end
 function make_hamiltonian2(n_states, υ, σ)
     mat = zeros(n_states, n_states)
     for i ∈ 1:n_states
-        mat[i,i] = (i / n_states) * υ
-        i < n_states ? mat[i,i+1] = σ : nothing 
-        println(i > 1 )
-        i > 1 ? (mat[i,i-1] = σ) : nothing 
+        mat[i, i] = (i / n_states) * υ
+        i < n_states ? mat[i, i + 1] = σ : nothing
+        println(i > 1)
+        i > 1 ? (mat[i, i - 1] = σ) : nothing
     end
     return mat
-end
+end

src/CategorizationModels.jl

@@ -1,25 +1,25 @@
 module CategorizationModels
 
-    using Distributions
-    using LinearAlgebra
+using Distributions
+using LinearAlgebra
 
-    import Distributions: ContinuousUnivariateDistribution
-    import Distributions: rand
-    import Distributions: logpdf
-        
-    export BayesianModel
-    export MarkovModel
-    export QuantumModel
-    export RationalModel
+import Distributions: ContinuousUnivariateDistribution
+import Distributions: rand
+import Distributions: logpdf
 
-    export predict
-    export logpdf 
-    export rand
-    export sumlogpdf
-    
-    include("common.jl")
-    include("rational.jl")
-    include("bayesian.jl")
-    include("markov.jl")
-    include("quantum.jl")
+export BayesianModel
+export MarkovModel
+export QuantumModel
+export RationalModel
+
+export predict
+export logpdf
+export rand
+export sumlogpdf
+
+include("common.jl")
+include("rational.jl")
+include("bayesian.jl")
+include("markov.jl")
+include("quantum.jl")
 end

src/bayesian.jl

@@ -14,9 +14,9 @@ A model object for the Bayesian model.
 - `λ_sk_s`: diffusion for stimulus s when s is evaluated first 
 - `n_states`: the number of evidence states
 """
-struct BayesianModel{T<:Real} <: Model 
+struct BayesianModel{T <: Real} <: Model
     μk::T
-    μs::T 
+    μs::T
     σk::T
     σs::T
     υ_ks_k::T
@@ -26,48 +26,46 @@ struct BayesianModel{T<:Real} <: Model
     n_states::Int
 end
 
-function BayesianModel(;μk,
-                        μs, 
-                        σk, 
-                        σs, 
-                        υ_ks_k, 
-                        υ_sk_s,
-                        λ_ks_k,
-                        λ_sk_s,
-                        n_states)
-
+function BayesianModel(; μk,
+    μs,
+    σk,
+    σs,
+    υ_ks_k,
+    υ_sk_s,
+    λ_ks_k,
+    λ_sk_s,
+    n_states)
     return BayesianModel(μk,
-                         μs, 
-                         σk, 
-                         σs, 
-                         υ_ks_k, 
-                         υ_sk_s,
-                         λ_ks_k,
-                         λ_sk_s,
-                         n_states)
+        μs,
+        σk,
+        σs,
+        υ_ks_k,
+        υ_sk_s,
+        λ_ks_k,
+        λ_sk_s,
+        n_states)
 end
 
 function BayesianModel(
     μk,
-    μs, 
-    σk, 
-    σs, 
-    υ_ks_k, 
+    μs,
+    σk,
+    σs,
+    υ_ks_k,
     υ_sk_s,
     λ_ks_k,
     λ_sk_s,
     n_states)
-
-    parms = promote(    
+    parms = promote(
         μk,
-        μs, 
-        σk, 
-        σs, 
-        υ_ks_k, 
+        μs,
+        σk,
+        σs,
+        υ_ks_k,
         υ_sk_s,
         λ_ks_k,
         λ_sk_s)
-        
+
     return BayesianModel(parms..., n_states)
 end
 
@@ -95,11 +93,11 @@ distribution of rating both stimuli in two orders. The joint distributions are a
 4. The joint probability distribution for k then s given stimulus s where element `pred[i,j]` is the probability of rating stimulus `s` as `i` and stimulus `k` as `j` 
 """
 function predict(model::BayesianModel{T}, n_options) where {T}
-    (;μk,μs,σk,σs,n_states) = model 
-    (;υ_ks_k, υ_sk_s,λ_ks_k,λ_sk_s) = model 
+    (; μk, μs, σk, σs, n_states) = model
+    (; υ_ks_k, υ_sk_s, λ_ks_k, λ_sk_s) = model
 
     n = div(n_states, n_options)
-    initial_states = Vector{Vector{T}}(undef,4)
+    initial_states = Vector{Vector{T}}(undef, 4)
     # initial state k then s for k stimulus
     initial_states[1] = compute_initial_state(model, μk, σk, n_states)
     # initial state s then k for s stimulus
@@ -110,7 +108,7 @@ function predict(model::BayesianModel{T}, n_options) where {T}
     # intensity matrix s then k for s stimulus 
     κ_sk_s = make_intensity_matrix(model, n_states, υ_sk_s)
 
-    transitions = Vector{Matrix{T}}(undef,4)
+    transitions = Vector{Matrix{T}}(undef, 4)
     # transition matrix k then s for k stimulus
     transitions[1] = exp(λ_ks_k * κ_ks_k)
     # transition matrix s then k for s stimulus 
@@ -122,27 +120,27 @@ function predict(model::BayesianModel{T}, n_options) where {T}
     initial_states[4] = transitions[3] * initial_states[3]
 
     # transition matrix s then k for k stimulus
-    transitions[2] = make_transition_matrix(model, 
-                                        initial_states[2], 
-                                        initial_states[1], 
-                                        transitions[1])
+    transitions[2] = make_transition_matrix(model,
+        initial_states[2],
+        initial_states[1],
+        transitions[1])
 
     # transition matrix k then s for s stimulus
-    transitions[4] = make_transition_matrix(model, 
-                                        initial_states[4], 
-                                        initial_states[3], 
-                                        transitions[3])
+    transitions[4] = make_transition_matrix(model,
+        initial_states[4],
+        initial_states[3],
+        transitions[3])
 
     # a projector for each option 
-    projectors = map(i -> 
-                        make_projector(n_states, n * (i - 1) + 1, i * n),
-                        1:n_options)
-                        
-    preds = map((s,t) -> 
-                    make_joint_dist(model, s, projectors, t),
-                    initial_states,
-                    transitions)
-    return preds 
+    projectors = map(i ->
+            make_projector(n_states, n * (i - 1) + 1, i * n),
+        1:n_options)
+
+    preds = map((s, t) ->
+            make_joint_dist(model, s, projectors, t),
+        initial_states,
+        transitions)
+    return preds
 end
 
 """
@@ -173,15 +171,15 @@ function make_intensity_matrix(model::BayesianModel, n_states, υ)
     mat = zeros(n_states, n_states)
     for c ∈ 1:n_states, r ∈ 1:n_states
         if r == (c - 1)
-            mat[r,c] = 1 - υ
+            mat[r, c] = 1 - υ
         elseif c == (r - 1)
-            mat[r,c] = 1 + υ
+            mat[r, c] = 1 + υ
         elseif r == c
-            mat[r,c] = -2
+            mat[r, c] = -2
         end
     end
-    mat[1,1] = -1 - υ
-    mat[end,end] = υ - 1
+    mat[1, 1] = -1 - υ
+    mat[end, end] = υ - 1
     return mat
 end
 
@@ -201,7 +199,7 @@ function make_transition_matrix(model::BayesianModel, s0_sk, s0_ks, T)
     n = length(s0_sk)
     probs = fill(0.0, n, n)
     for k ∈ 1:n, j ∈ 1:n
-        probs[j,k] = s0_ks[j] * T[k,j] / s0_sk[k]
+        probs[j, k] = s0_ks[j] * T[k, j] / s0_sk[k]
     end
-    return probs 
-end
+    return probs
+end