Cleanup

Project.toml

@@ -16,15 +16,12 @@ Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 DocStringExtensions = "ffbed154-4ef7-542d-bbb7-c09d3a79fcae"
 EffectSizes = "e248de7e-9197-5860-972e-353a2af44d75"
 ElasticArrays = "fdbdab4c-e67f-52f5-8c3f-e7b388dad3d4"
-FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
 FittedItemBanks = "3f797b09-34e4-41d7-acf6-3302ae3248a5"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 HypothesisTests = "09f84164-cd44-5f33-b23f-e6b0d136a0d5"
-Interpolations = "a98d9a8b-a2ab-59e6-89dd-64a1c18fca59"
 Lazy = "50d2b5c4-7a5e-59d5-8109-a42b560f39c0"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 LogarithmicNumbers = "aa2f6b4e-9042-5d33-9679-40d3a6b85899"
-MacroTools = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"
 Mmap = "a63ad114-7e13-5084-954f-fe012c677804"
 PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
 PrettyPrinting = "54e16d92-306c-5ea0-a30b-337be88ac337"
@@ -33,8 +30,6 @@ PsychometricsBazaarBase = "b0d9cada-d963-45e9-a4c6-4746243987f1"
 QuickHeaps = "30b38841-0f52-47f8-a5f8-18d5d4064379"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
-Setfield = "efcf1570-3423-57d1-acb7-fd33fddbac46"
-SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
 StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
 StatsFuns = "4c63d2b9-4356-54db-8cca-17b64c39e42c"
@@ -55,15 +50,12 @@ Distributions = "^0.25.88"
 DocStringExtensions = " ^0.9"
 EffectSizes = "^1.0.1"
 ElasticArrays = "1.2.12"
-FillArrays = "0.13, 1.5.0"
 FittedItemBanks = "^0.7.3"
 ForwardDiff = "1"
 HypothesisTests = "^0.10.12, ^0.11.0"
-Interpolations = "^0.14, ^0.15"
 Lazy = "0.15"
 LinearAlgebra = "^1.11"
 LogarithmicNumbers = "1"
-MacroTools = "^0.5.6"
 Mmap = "^1.11"
 PrecompileTools = "1.2.1"
 PrettyPrinting = "0.4.2"
@@ -72,8 +64,6 @@ PsychometricsBazaarBase = "^0.8.7"
 QuickHeaps = "0.2.2"
 Random = "^1.11"
 Reexport = "1"
-Setfield = "^1"
-SparseArrays = "^1.11"
 StaticArrays = "1"
 StatsBase = "^0.34"
 StatsFuns = "^0.9.15, ^1"

src/Compat/CatR.jl

@@ -61,23 +61,6 @@ end
 
 const ability_estimator_aliases = _ability_estimator_aliases()
 
-#=
-        for (resp_exp, resp_exp_nick) in resp_exp_nick_pairs
-            next_item_rule = NextItemRule(
-                ExpectationBasedItemCriterion(resp_exp, AbilityVariance(numtools.integrator, distribution_estimator(abil_est)))
-            )
-            next_item_rule = preallocate(next_item_rule)
-            est_next_item_rule_pairs[Symbol("$(abil_est_str)_mepv_$(resp_exp_nick)")] = (abil_est, next_item_rule)
-            next_item_rule = NextItemRule(
-                ExpectationBasedItemCriterion(resp_exp, InformationItemCriterion(abil_est))
-            )
-            next_item_rule = preallocate(next_item_rule)
-            est_next_item_rule_pairs[Symbol("$(abil_est_str)_mei_$(resp_exp_nick)")] = (abil_est, next_item_rule)
-        end
-        est_next_item_rule_pairs[Symbol("$(abil_est_str)_mi")] = (abil_est, InformationItemCriterion(abil_est))
-=#
-
-
 function setup_integrator(lo=-4.0, hi=4.0, pts=33)
     Integrators.MidpointIntegrator(range(lo, hi, pts))
 end

src/NextItemRules/NextItemRules.jl

@@ -64,6 +64,12 @@ export ObservedInformationPointwiseItemCriterion
 export RawEmpiricalInformationPointwiseItemCriterion
 export EmpiricalInformationPointwiseItemCriterion
 
+public PointwiseNextItemRule, PointwiseFirstNextItemRule
+public WeightedStateMultiCriterion, WeightedItemMultiCriterion
+public GreedyForcedContentBalancer
+public PosteriorExpectedKLInformationItemCriterion
+public alt_expected_1d_item_information, alt_expected_mirt_item_information, alt_expected_mirt_3pl_item_information
+
 # Prelude
 include("./prelude/abstract.jl")
 include("./prelude/next_item_rule.jl")

src/NextItemRules/criteria/pointwise/information_support.jl

@@ -30,42 +30,6 @@ function log_resp(ir::ItemResponse{<:CdfMirtItemBank}, val, θ)
     end
 end
 
-#=
-# XXX: Not sure if this is optimal numerically or speed wise -- possibly it
-# would be better to just transform to linear space in this case?
-@inline function log_transform_irf_y(guess, slip, y)
-    # log space version of guess + irf_size(guess, slip) * y
-    logaddexp(log(guess), log(irf_size(guess, slip)) + y)
-end
-
-@inline function log_transform_irf_y(ir::ItemResponse{<:GuessItemBank}, response, y)
-    guess = y_offset(ir.item_bank, ir.index)
-    if response
-        log_transform_irf_y(guess, 0.0, y)
-    else
-        log_transform_irf_y(0.0, guess, y)
-    end
-end
-
-@inline function log_transform_irf_y(ir::ItemResponse{<:SlipItemBank}, response, y)
-    slip = y_offset(ir.item_bank, ir.index)
-    if response
-        log_transform_irf_y(0.0, slip, y)
-    else
-        log_transform_irf_y(slip, 0.0, y)
-    end
-end
-
-function log_resp_vec(ir::ItemResponse{<:AnySlipOrGuessItemBank}, θ)
-    r = log_resp_vec(inner_item_response(ir), θ)
-    SVector(log_transform_irf_y(ir, false, r[1]), log_transform_irf_y(ir, true, r[2]))
-end
-
-function log_resp(ir::ItemResponse{<:AnySlipOrGuessItemBank}, val, θ)
-    log_transform_irf_y(ir, val, log_resp(inner_item_response(ir), val, θ))
-end
-=#
-
 log_resp(ir::ItemResponse{<:GuessAndSlipItemBank}, response, θ) = log(resp(ir, response, θ))
 log_resp(ir::ItemResponse{<:GuessAndSlipItemBank}, θ) = log(resp(ir, θ))
 log_resp_vec(ir::ItemResponse{<:GuessAndSlipItemBank}, θ) = log.(resp_vec(ir, θ))

src/NextItemRules/criteria/pointwise/kl.jl

@@ -12,6 +12,9 @@ struct PosteriorExpectedKLInformationItemCriterion{
     DistributionEstimatorT <: DistributionAbilityEstimator,
     IntegratorT <: AbilityIntegrator
 } <: PointwiseItemCriterion
+    point_estimator::PointEstimatorT
+    distribution_estimator::DistributionEstimatorT
+    integrator::IntegratorT
 end
 
 function PosteriorExpectedKLInformationItemCriterion(bits...)