rename AbstractHybridCase to AbstractHybridProblem (#15)

Author: bgctw

dev/doubleMM.jl

@@ -16,24 +16,24 @@ using OptimizationOptimisers
 using Bijectors
 using UnicodePlots
 
-const case = DoubleMM.DoubleMMCase()
+const prob = DoubleMM.DoubleMMCase()
 scenario = (:default,)
 rng = StableRNG(111)
 
-par_templates = get_hybridcase_par_templates(case; scenario)
+par_templates = get_hybridproblem_par_templates(prob; scenario)
 
-#n_covar = get_hybridcase_n_covar(case; scenario)
-#, n_batch, n_θM, n_θP) = get_hybridcase_sizes(case; scenario)
+#n_covar = get_hybridproblem_n_covar(prob; scenario)
+#, n_batch, n_θM, n_θP) = get_hybridproblem_sizes(prob; scenario)
 
 (; xM, n_site, θP_true, θMs_true, xP, y_global_true, y_true, y_global_o, y_o, y_unc
-) = gen_hybridcase_synthetic(rng, case; scenario);
+) = gen_hybridcase_synthetic(rng, prob; scenario);
 
 n_covar = size(xM,1)
 
 
 #----- fit g to θMs_true
-g, ϕg0 = get_hybridcase_MLapplicator(case; scenario);
-(; transP, transM) = get_hybridcase_transforms(case; scenario)
+g, ϕg0 = get_hybridproblem_MLapplicator(prob; scenario);
+(; transP, transM) = get_hybridproblem_transforms(prob; scenario)
 
 function loss_g(ϕg, x, g, transM)
     ζMs = g(x, ϕg) # predict the log of the parameters
@@ -52,8 +52,8 @@ res = Optimization.solve(optprob, Adam(0.02), callback = callback_loss(100), max
 l1, θMs_pred = loss_g(ϕg_opt1, xM, g, transM)
 scatterplot(vec(θMs_true), vec(θMs_pred))
 
-f = get_hybridcase_PBmodel(case; scenario)
-py = get_hybridcase_neg_logden_obs(case; scenario)
+f = get_hybridproblem_PBmodel(prob; scenario)
+py = get_hybridproblem_neg_logden_obs(prob; scenario)
 
 #----------- fit g and θP to y_o
 () -> begin
@@ -85,8 +85,8 @@ end
 
 #---------- HVI
 n_MC = 3
-(; transP, transM) = get_hybridcase_transforms(case; scenario)
-FT = get_hybridcase_float_type(case; scenario)
+(; transP, transM) = get_hybridproblem_transforms(prob; scenario)
+FT = get_hybridproblem_float_type(prob; scenario)
 
 (; ϕ, transPMs_batch, interpreters, get_transPMs, get_ca_int_PMs) = init_hybrid_params(
     θP_true, θMs_true[:, 1], ϕg_opt1, n_batch; transP, transM);
@@ -167,7 +167,7 @@ mean_σ_o_MC = 0.006042
 ϕ = CA.getdata(ϕ_ini) |> Flux.gpu;
 xM_gpu = xM |> Flux.gpu;
 scenario_flux = (scenario..., :use_Flux)
-g_flux, _ = get_hybridcase_MLapplicator(case; scenario = scenario_flux);
+g_flux, _ = get_hybridproblem_MLapplicator(prob; scenario = scenario_flux);
 
 # otpimize using LUX
 () -> begin
@@ -205,7 +205,7 @@ gr = Zygote.gradient(fcost,
 gr_c = CA.ComponentArray(gr[1] |> Flux.cpu, CA.getaxes(ϕ_ini)...)
 
 train_loader = MLUtils.DataLoader((xM_gpu, xP, y_o, y_unc), batchsize = n_batch)
-#train_loader = get_hybridcase_train_dataloader(case, rng; scenario = (scenario..., :use_Flux))
+#train_loader = get_hybridproblem_train_dataloader(prob, rng; scenario = (scenario..., :use_Flux))
 
 optf = Optimization.OptimizationFunction(
     (ϕ, data) -> begin

ext/HybridVariationalInferenceFluxExt.jl

@@ -36,13 +36,13 @@ end
 # end
 
 function HVI.construct_3layer_MLApplicator(
-        rng::AbstractRNG, case::HVI.AbstractHybridCase, ::Val{:Flux};
+        rng::AbstractRNG, prob::HVI.AbstractHybridProblem, ::Val{:Flux};
         scenario::NTuple = ())
-    (;θM) = get_hybridcase_par_templates(case; scenario)
+    (;θM) = get_hybridproblem_par_templates(prob; scenario)
     n_out = length(θM)
-    n_covar = get_hybridcase_n_covar(case; scenario)
-    #(; n_covar, n_θM) = get_hybridcase_sizes(case; scenario)
-    float_type = get_hybridcase_float_type(case; scenario)
+    n_covar = get_hybridproblem_n_covar(prob; scenario)
+    #(; n_covar, n_θM) = get_hybridproblem_sizes(prob; scenario)
+    float_type = get_hybridproblem_float_type(prob; scenario)
     is_using_dropout = :use_dropout ∈ scenario
     is_using_dropout && error("dropout scenario not supported with Flux yet.")
     g_chain = Flux.Chain(

ext/HybridVariationalInferenceSimpleChainsExt.jl

@@ -20,11 +20,11 @@ end
 HVI.apply_model(app::SimpleChainsApplicator, x, ϕ) = app.m(x, ϕ)
 
 function HVI.construct_3layer_MLApplicator(
-    rng::AbstractRNG, case::HVI.AbstractHybridCase, ::Val{:SimpleChains};
+    rng::AbstractRNG, prob::HVI.AbstractHybridProblem, ::Val{:SimpleChains};
     scenario::NTuple = ())
-    n_covar = get_hybridcase_n_covar(case; scenario)
-    FloatType = get_hybridcase_float_type(case; scenario)
-    (;θM) = get_hybridcase_par_templates(case; scenario)
+    n_covar = get_hybridproblem_n_covar(prob; scenario)
+    FloatType = get_hybridproblem_float_type(prob; scenario)
+    (;θM) = get_hybridproblem_par_templates(prob; scenario)
     n_out = length(θM)
     is_using_dropout = :use_dropout ∈ scenario
     g_chain = if is_using_dropout

src/hybrid_case.jl src/AbstractHybridProblem.jlsrc/hybrid_case.jl renamed to src/AbstractHybridProblem.jl

@@ -2,40 +2,40 @@
 Type to dispatch constructing data and network structures
 for different cases of hybrid problem setups
 
-For a specific case, provide functions that specify details
-- `get_hybridcase_MLapplicator`
-- `get_hybridcase_PBmodel`
-- `get_hybridcase_neg_logden_obs`
-- `get_hybridcase_par_templates`
-- `get_hybridcase_transforms`
-- `get_hybridcase_train_dataloader` (default depends on `gen_hybridcase_synthetic`)
+For a specific prob, provide functions that specify details
+- `get_hybridproblem_MLapplicator`
+- `get_hybridproblem_PBmodel`
+- `get_hybridproblem_neg_logden_obs`
+- `get_hybridproblem_par_templates`
+- `get_hybridproblem_transforms`
+- `get_hybridproblem_train_dataloader` (default depends on `gen_hybridcase_synthetic`)
 optionally
 - `gen_hybridcase_synthetic`
-- `get_hybridcase_n_covar` (defaults to number of rows in xM in train_dataloader )
-- `get_hybridcase_float_type` (defaults to `eltype(θM)`)
-- `get_hybridcase_cor_starts` (defaults to include all correlations: `(P=(1,), M=(1,))`)
+- `get_hybridproblem_n_covar` (defaults to number of rows in xM in train_dataloader )
+- `get_hybridproblem_float_type` (defaults to `eltype(θM)`)
+- `get_hybridproblem_cor_starts` (defaults to include all correlations: `(P=(1,), M=(1,))`)
 """
-abstract type AbstractHybridCase end;
+abstract type AbstractHybridProblem end;
 
 
 """
-    get_hybridcase_MLapplicator([rng::AbstractRNG,] ::AbstractHybridCase; scenario=())
+    get_hybridproblem_MLapplicator([rng::AbstractRNG,] ::AbstractHybridProblem; scenario=())
 
-Construct the machine learning model fro given problem case and ML-Framework and 
+Construct the machine learning model fro given problem prob and ML-Framework and 
 scenario.
 
 returns a Tuple of
 - AbstractModelApplicator
 - initial parameter vector
 """
-function get_hybridcase_MLapplicator end    
+function get_hybridproblem_MLapplicator end    
 
-function get_hybridcase_MLapplicator(case::AbstractHybridCase; scenario=())
-    get_hybridcase_MLapplicator(Random.default_rng(), case; scenario)
+function get_hybridproblem_MLapplicator(prob::AbstractHybridProblem; scenario=())
+    get_hybridproblem_MLapplicator(Random.default_rng(), prob; scenario)
 end
 
 """
-    get_hybridcase_PBmodel(::AbstractHybridCase; scenario::NTuple=())
+    get_hybridproblem_PBmodel(::AbstractHybridProblem; scenario::NTuple=())
 
 Construct the process-based model function 
 `f(θP::AbstractVector, θMs::AbstractMatrix, x) -> (AbstractVector, AbstractMatrix)`
@@ -48,59 +48,59 @@ returns a tuple of predictions with components
 - first, those that are constant across sites
 - second, those that vary across sites, with a column for each site
 """
-function get_hybridcase_PBmodel end
+function get_hybridproblem_PBmodel end
 
 """
-    get_hybridcase_neg_logden_obs(::AbstractHybridCase; scenario)
+    get_hybridproblem_neg_logden_obs(::AbstractHybridProblem; scenario)
 
 Provide a `function(y_obs, ypred) -> Real` that computes the negative logdensity
 of the observations, given the predictions.
 """
-function get_hybridcase_neg_logden_obs end    
+function get_hybridproblem_neg_logden_obs end    
 
 
 """
-    get_hybridcase_par_templates(::AbstractHybridCase; scenario)
+    get_hybridproblem_par_templates(::AbstractHybridProblem; scenario)
 
 Provide tuple of templates of ComponentVectors `θP` and `θM`.
 """
-function get_hybridcase_par_templates end    
+function get_hybridproblem_par_templates end    
 
 
 """
-    get_hybridcase_transforms(::AbstractHybridCase; scenario)
+    get_hybridproblem_transforms(::AbstractHybridProblem; scenario)
 
 Return a NamedTupe of
 - `transP`: Bijectors.Transform for the global PBM parameters, θP
 - `transM`: Bijectors.Transform for the single-site PBM parameters, θM
 """
-function get_hybridcase_transforms end
+function get_hybridproblem_transforms end
 
 # """
-#     get_hybridcase_par_templates(::AbstractHybridCase; scenario)
+#     get_hybridproblem_par_templates(::AbstractHybridProblem; scenario)
 # Provide a NamedTuple of number of 
 # - n_covar: covariates xM
 # - n_site: all sites in the data
 # - n_batch: sites in one minibatch during fitting
 # - n_θM, n_θP: entries in parameter vectors
 # """
-# function get_hybridcase_sizes end
+# function get_hybridproblem_sizes end
 
 """
-    get_hybridcase_n_covar(::AbstractHybridCase; scenario)
+    get_hybridproblem_n_covar(::AbstractHybridProblem; scenario)
 
 Provide the number of covariates. Default returns the number of rows in `xM` from
-`get_hybridcase_train_dataloader`.
+`get_hybridproblem_train_dataloader`.
 """
-function get_hybridcase_n_covar(case::AbstractHybridCase; scenario)
-    train_loader = get_hybridcase_train_dataloader(Random.default_rng(), case; scenario)
+function get_hybridproblem_n_covar(prob::AbstractHybridProblem; scenario)
+    train_loader = get_hybridproblem_train_dataloader(Random.default_rng(), prob; scenario)
     (xM, xP, y_o, y_unc) = first(train_loader)
     n_covar = size(xM, 1)
     return(n_covar)
 end
 
 """
-    gen_hybridcase_synthetic([rng,] ::AbstractHybridCase; scenario)
+    gen_hybridcase_synthetic([rng,] ::AbstractHybridProblem; scenario)
 
 Setup synthetic data, a NamedTuple of
 - xM: matrix of covariates, with one column per site
@@ -114,40 +114,40 @@ Setup synthetic data, a NamedTuple of
 function gen_hybridcase_synthetic end
 
 """
-    get_hybridcase_float_type(::AbstractHybridCase; scenario)
+    get_hybridproblem_float_type(::AbstractHybridProblem; scenario)
 
 Determine the FloatType for given Case and scenario, defaults to Float32
 """
-function get_hybridcase_float_type(case::AbstractHybridCase; scenario=())
-    return eltype(get_hybridcase_par_templates(case; scenario).θM)
+function get_hybridproblem_float_type(prob::AbstractHybridProblem; scenario=())
+    return eltype(get_hybridproblem_par_templates(prob; scenario).θM)
 end
 
 """
-    get_hybridcase_train_dataloader([rng,] ::AbstractHybridCase; scenario)
+    get_hybridproblem_train_dataloader([rng,] ::AbstractHybridProblem; scenario)
 
 Return a DataLoader that provides a tuple of
 - `xM`: matrix of covariates, with one column per site
 - `xP`: Iterator of process-model drivers, with one element per site
 - `y_o`: matrix of observations with added noise, with one column per site
 - `y_unc`: matrix `sizeof(y_o)` of uncertainty information 
 """
-function get_hybridcase_train_dataloader(rng::AbstractRNG, case::AbstractHybridCase; 
+function get_hybridproblem_train_dataloader(rng::AbstractRNG, prob::AbstractHybridProblem; 
     scenario = ())
-    (; xM, xP, y_o, y_unc) = gen_hybridcase_synthetic(rng, case; scenario)
+    (; xM, xP, y_o, y_unc) = gen_hybridcase_synthetic(rng, prob; scenario)
     n_batch = 10
     xM_gpu = :use_Flux ∈ scenario ? CuArray(xM) : xM
     train_loader = MLUtils.DataLoader((xM_gpu, xP, y_o, y_unc), batchsize = n_batch)
     return(train_loader)
 end
 
-function get_hybridcase_train_dataloader(case::AbstractHybridCase; scenario = ())
+function get_hybridproblem_train_dataloader(prob::AbstractHybridProblem; scenario = ())
     rng::AbstractRNG = Random.default_rng()
-    get_hybridcase_train_dataloader(rng, case; scenario)
+    get_hybridproblem_train_dataloader(rng, prob; scenario)
 end
 
 
 """
-    get_hybridcase_cor_starts(case::AbstractHybridCase; scenario)
+    get_hybridproblem_cor_starts(prob::AbstractHybridProblem; scenario)
 
 Specify blocks in correlation matrices among parameters.
 Returns a NamedTuple.
@@ -163,7 +163,7 @@ then the first subrange starts at position 1 and the second subrange starts at p
 If there is only single block of all ML-predicted parameters being correlated 
 with each other then this block starts at position 1: `(P=(1,3), M=(1,))`.
 """
-function get_hybridcase_cor_starts(case::AbstractHybridCase; scenario = ())
+function get_hybridproblem_cor_starts(prob::AbstractHybridProblem; scenario = ())
     (P=(1,), M=(1,))
 end
 

src/ComponentArrayInterpreter.jl

@@ -137,7 +137,7 @@ function ComponentArrayInterpreter(
     ComponentArrayInterpreter(axes_ext)
 end
 
-# ambuiguity with two empty Tuples (edge case that does not make sense)
+# ambuiguity with two empty Tuples (edge prob that does not make sense)
 # Empty ComponentVector with no other array dimensions -> empty componentVector
 function ComponentArrayInterpreter(n_dims1::Tuple{}, n_dims2::Tuple{})
     ComponentArrayInterpreter(CA.ComponentVector())

src/DoubleMM/f_doubleMM.jl

@@ -1,4 +1,4 @@
-struct DoubleMMCase <: AbstractHybridCase end
+struct DoubleMMCase <: AbstractHybridProblem end
 
 
 θP = CA.ComponentVector{Float32}(r0 = 0.3, K2 = 2.0)
@@ -18,19 +18,19 @@ function f_doubleMM(θ::AbstractVector, x)
     return (y)
 end
 
-function HVI.get_hybridcase_par_templates(::DoubleMMCase; scenario::NTuple = ())
+function HVI.get_hybridproblem_par_templates(::DoubleMMCase; scenario::NTuple = ())
     (; θP, θM)
 end
 
-function HVI.get_hybridcase_transforms(::DoubleMMCase; scenario::NTuple = ())
+function HVI.get_hybridproblem_transforms(::DoubleMMCase; scenario::NTuple = ())
     (; transP, transM)
 end
 
-function HVI.get_hybridcase_neg_logden_obs(::DoubleMMCase; scenario::NTuple = ())
+function HVI.get_hybridproblem_neg_logden_obs(::DoubleMMCase; scenario::NTuple = ())
     neg_logden_indep_normal
 end
 
-# function HVI.get_hybridcase_sizes(::DoubleMMCase; scenario = ())
+# function HVI.get_hybridproblem_sizes(::DoubleMMCase; scenario = ())
 #     n_covar_pc = 2
 #     n_covar = n_covar_pc + 3 # linear dependent
 #     #n_site = 10^n_covar_pc
@@ -41,7 +41,7 @@ end
 #     (; n_covar, n_batch, n_θM, n_θP)
 # end
 
-function HVI.get_hybridcase_PBmodel(::DoubleMMCase; scenario::NTuple = ())
+function HVI.get_hybridproblem_PBmodel(::DoubleMMCase; scenario::NTuple = ())
     #fsite = (θ, x_site) -> f_doubleMM(θ)  # omit x_site drivers
     function f_doubleMM_with_global(θP::AbstractVector, θMs::AbstractMatrix, x)
         pred_sites = applyf(f_doubleMM, θMs, θP, x)
@@ -50,26 +50,26 @@ function HVI.get_hybridcase_PBmodel(::DoubleMMCase; scenario::NTuple = ())
     end
 end
 
-# function HVI.get_hybridcase_float_type(::DoubleMMCase; scenario)
+# function HVI.get_hybridproblem_float_type(::DoubleMMCase; scenario)
 #     return Float32
 # end
 
 const xP_S1 = Float32[1.0, 1.0, 1.0, 1.0, 0.4, 0.3, 0.1]
 const xP_S2 = Float32[1.0, 3.0, 4.0, 5.0, 5.0, 5.0, 5.0]
 
-function HVI.gen_hybridcase_synthetic(rng::AbstractRNG, case::DoubleMMCase;
+function HVI.gen_hybridcase_synthetic(rng::AbstractRNG, prob::DoubleMMCase;
         scenario = ())
     n_covar_pc = 2
     n_site = 200
     n_covar = 5
     n_θM = length(θM)
-    FloatType = get_hybridcase_float_type(case; scenario)
+    FloatType = get_hybridproblem_float_type(prob; scenario)
     xM, θMs_true0 = gen_cov_pred(rng, FloatType, n_covar_pc, n_covar, n_site, n_θM;
         rhodec = 8, is_using_dropout = false)
     int_θMs_sites = ComponentArrayInterpreter(θM, (n_site,))
     # normalize to be distributed around the prescribed true values
     θMs_true = int_θMs_sites(scale_centered_at(θMs_true0, θM, FloatType(0.1)))
-    f = get_hybridcase_PBmodel(case; scenario)
+    f = get_hybridproblem_PBmodel(prob; scenario)
     xP = fill((;S1=xP_S1, S2=xP_S2), n_site)
     y_global_true, y_true = f(θP, θMs_true, xP)
     σ_o = FloatType(0.01)
@@ -91,10 +91,10 @@ function HVI.gen_hybridcase_synthetic(rng::AbstractRNG, case::DoubleMMCase;
     )
 end
 
-function HVI.get_hybridcase_MLapplicator(
-    rng::AbstractRNG, case::HVI.DoubleMM.DoubleMMCase; scenario = ())
+function HVI.get_hybridproblem_MLapplicator(
+    rng::AbstractRNG, prob::HVI.DoubleMM.DoubleMMCase; scenario = ())
     ml_engine = select_ml_engine(; scenario)
-    construct_3layer_MLApplicator(rng, case, ml_engine; scenario)
+    construct_3layer_MLApplicator(rng, prob, ml_engine; scenario)
 end