Defer inclusion of generated quantity code (#364)

NAMESPACE

@@ -86,9 +86,15 @@ S3method(coalesceGridTime,GridPrediction)
 S3method(coalesceGridTime,default)
 S3method(compileStanModel,JointModel)
 S3method(dim,Quantities)
+S3method(enableGQ,JointModel)
+S3method(enableGQ,LongitudinalGSF)
+S3method(enableGQ,LongitudinalRandomSlope)
+S3method(enableGQ,LongitudinalSteinFojo)
+S3method(enableGQ,default)
 S3method(enableLink,LongitudinalGSF)
 S3method(enableLink,LongitudinalRandomSlope)
 S3method(enableLink,LongitudinalSteinFojo)
+S3method(enableLink,default)
 S3method(extractVariableNames,DataSubject)
 S3method(extractVariableNames,DataSurvival)
 S3method(generateQuantities,JointModelSamples)
@@ -202,6 +208,7 @@ export(as_stan_list)
 export(autoplot)
 export(brierScore)
 export(compileStanModel)
+export(enableGQ)
 export(enableLink)
 export(generateQuantities)
 export(getParameters)

R/JointModel.R

@@ -80,16 +80,24 @@ JointModel <- function(
 }
 
 
+#' @export
+enableGQ.JointModel <- function(object, ...) {
+    merge(
+        enableGQ(object@survival),
+        enableGQ(object@longitudinal)
+    )
+}
+
+
 #' `JointModel` -> `StanModule`
 #'
 #' Converts a [`JointModel`] object to a [`StanModule`] object
 #'
 #' @inheritParams JointModel-Shared
-#' @param include_gq (`logical`)\cr whether to include the generated quantities block.
 #' @family JointModel
 #' @family as.StanModule
 #' @export
-as.StanModule.JointModel <- function(object, include_gq = FALSE, ...) {
+as.StanModule.JointModel <- function(object, ...) {
 
     base_model <- read_stan("base/base.stan")
 
@@ -113,9 +121,6 @@ as.StanModule.JointModel <- function(object, include_gq = FALSE, ...) {
         StanModule(stan_full)
     )
 
-    if (!include_gq) {
-        x@generated_quantities <- ""
-    }
     return(x)
 }
 

R/JointModelSamples.R

@@ -52,9 +52,13 @@ generateQuantities.JointModelSamples <- function(object, generator, type, ...) {
         ) |>
         StanModule()
 
-    stanobj <- merge(
-        as.StanModule(object@model, include_gq = TRUE),
-        quant_stanobj
+    stanobj <- Reduce(
+        merge,
+        list(
+            as.StanModule(object@model),
+            enableGQ(object@model),
+            quant_stanobj
+        )
     )
 
     model <- compileStanModel(stanobj)

R/LongitudinalGSF.R

@@ -126,6 +126,13 @@ LongitudinalGSF <- function(
 }
 
 
+
+#' @export
+enableGQ.LongitudinalGSF <- function(object, ...) {
+    StanModule("lm-gsf/quantities.stan")
+}
+
+
 #' @export
 enableLink.LongitudinalGSF <- function(object, ...) {
     object@stan <- merge(

R/LongitudinalRandomSlope.R

@@ -60,6 +60,11 @@ LongitudinalRandomSlope <- function(
 }
 
 
+#' @export
+enableGQ.LongitudinalRandomSlope <- function(object, ...) {
+    StanModule("lm-random-slope/quantities.stan")
+}
+
 #' @export
 enableLink.LongitudinalRandomSlope <- function(object, ...) {
     object@stan <- merge(

R/LongitudinalSteinFojo.R

@@ -111,6 +111,11 @@ LongitudinalSteinFojo <- function(
 
 
 
+#' @export
+enableGQ.LongitudinalSteinFojo <- function(object, ...) {
+    StanModule("lm-stein-fojo/quantities.stan")
+}
+
 #' @export
 enableLink.LongitudinalSteinFojo <- function(object, ...) {
     object@stan <- merge(

R/generics.R

@@ -410,11 +410,32 @@ resolvePromise.default <- function(object, ...) {
 enableLink <- function(object, ...) {
     UseMethod("enableLink")
 }
+#' @export
 enableLink.default <- function(object, ...) {
     object
 }
 
 
+#' Enable Generated Quantities Generic
+#'
+#' @param object ([`StanModel`])\cr to enable generated quantities for.
+#' @param ... Not used.
+#'
+#' Optional hook method that is called on a [`StanModel`] if attempting to use
+#' either [`LongitudinalQuantities`] or [`SurvivalQuantities`]
+#'
+#' @return [`StanModule`] object
+#'
+#' @export
+enableGQ <- function(object, ...) {
+    UseMethod("enableGQ")
+}
+#' @export
+enableGQ.default <- function(object, ...) {
+    StanModule()
+}
+
+
 
 #' Get Prediction Names
 #'

_pkgdown.yml

@@ -168,6 +168,7 @@ reference:
       - sampleObservations
       - sampleSubjects
       - enableLink
+      - enableGQ
       - as_formula
       - getPredictionNames
 

inst/stan/lm-gsf/functions.stan

@@ -21,15 +21,5 @@ functions {
         );
         return result;
     }
-
-    vector lm_predict_value(vector time, matrix long_gq_parameters) {
-        return sld(
-            time,
-            long_gq_parameters[,1],
-            long_gq_parameters[,2],
-            long_gq_parameters[,3],
-            long_gq_parameters[,4]
-        );
-    }
 }
 

inst/stan/lm-gsf/model.stan

@@ -94,21 +94,3 @@ model {
     lm_gsf_psi_phi ~ beta(lm_gsf_a_phi[subject_arm_index], lm_gsf_b_phi[subject_arm_index]);
 }
 
-
-generated quantities {
-    //
-    // Source - lm-gsf/model.stan
-    //
-    matrix[n_subjects, 4] long_gq_parameters;
-    long_gq_parameters[, 1] = lm_gsf_psi_bsld;
-    long_gq_parameters[, 2] = lm_gsf_psi_ks;
-    long_gq_parameters[, 3] = lm_gsf_psi_kg;
-    long_gq_parameters[, 4] = lm_gsf_psi_phi;
-
-
-    matrix[gq_n_quant, 4] long_gq_pop_parameters;
-    long_gq_pop_parameters[, 1] = exp(lm_gsf_mu_bsld[gq_long_pop_study_index]);
-    long_gq_pop_parameters[, 2] = exp(lm_gsf_mu_ks[gq_long_pop_arm_index]);
-    long_gq_pop_parameters[, 3] = exp(lm_gsf_mu_kg[gq_long_pop_arm_index]);
-    long_gq_pop_parameters[, 4] = lm_gsf_a_phi[gq_long_pop_arm_index] ./ (lm_gsf_a_phi[gq_long_pop_arm_index] + lm_gsf_b_phi[gq_long_pop_arm_index]);
-}

inst/stan/lm-gsf/quantities.stan

@@ -0,0 +1,29 @@
+functions {
+    vector lm_predict_value(vector time, matrix long_gq_parameters) {
+        return sld(
+            time,
+            long_gq_parameters[,1],
+            long_gq_parameters[,2],
+            long_gq_parameters[,3],
+            long_gq_parameters[,4]
+        );
+    }
+}
+
+generated quantities {
+    //
+    // Source - lm-gsf/quantities.stan
+    //
+    matrix[n_subjects, 4] long_gq_parameters;
+    long_gq_parameters[, 1] = lm_gsf_psi_bsld;
+    long_gq_parameters[, 2] = lm_gsf_psi_ks;
+    long_gq_parameters[, 3] = lm_gsf_psi_kg;
+    long_gq_parameters[, 4] = lm_gsf_psi_phi;
+
+
+    matrix[gq_n_quant, 4] long_gq_pop_parameters;
+    long_gq_pop_parameters[, 1] = exp(lm_gsf_mu_bsld[gq_long_pop_study_index]);
+    long_gq_pop_parameters[, 2] = exp(lm_gsf_mu_ks[gq_long_pop_arm_index]);
+    long_gq_pop_parameters[, 3] = exp(lm_gsf_mu_kg[gq_long_pop_arm_index]);
+    long_gq_pop_parameters[, 4] = lm_gsf_a_phi[gq_long_pop_arm_index] ./ (lm_gsf_a_phi[gq_long_pop_arm_index] + lm_gsf_b_phi[gq_long_pop_arm_index]);
+}

inst/stan/lm-random-slope/model.stan

@@ -1,15 +1,4 @@
 
-functions {
-    //
-    // Source - lm-random-slope/model.stan
-    //
-    vector lm_predict_value(vector time, matrix long_gq_parameters) {
-        int nrow = rows(time);
-        return (
-            long_gq_parameters[, 1] + long_gq_parameters[, 2] .* time
-        );
-    }
-}
 
 
 parameters {
@@ -61,16 +50,3 @@ model {
 }
 
 
-generated quantities {
-    //
-    // Source - lm-random-slope/model.stan
-    //
-    matrix[n_subjects, 2] long_gq_parameters;
-    long_gq_parameters[, 1] = lm_rs_ind_intercept;
-    long_gq_parameters[, 2] = lm_rs_ind_rnd_slope;
-
-
-    matrix[gq_n_quant, 2] long_gq_pop_parameters;
-    long_gq_pop_parameters[, 1] = to_vector(lm_rs_intercept[gq_long_pop_study_index]);
-    long_gq_pop_parameters[, 2] = to_vector(lm_rs_slope_mu[gq_long_pop_arm_index]);
-}

inst/stan/lm-random-slope/quantities.stan

@@ -0,0 +1,26 @@
+
+functions {
+    //
+    // Source - lm-random-slope/quantities.stan
+    //
+    vector lm_predict_value(vector time, matrix long_gq_parameters) {
+        int nrow = rows(time);
+        return (
+            long_gq_parameters[, 1] + long_gq_parameters[, 2] .* time
+        );
+    }
+}
+
+generated quantities {
+    //
+    // Source - lm-random-slope/quantities.stan
+    //
+    matrix[n_subjects, 2] long_gq_parameters;
+    long_gq_parameters[, 1] = lm_rs_ind_intercept;
+    long_gq_parameters[, 2] = lm_rs_ind_rnd_slope;
+
+
+    matrix[gq_n_quant, 2] long_gq_pop_parameters;
+    long_gq_pop_parameters[, 1] = to_vector(lm_rs_intercept[gq_long_pop_study_index]);
+    long_gq_pop_parameters[, 2] = to_vector(lm_rs_slope_mu[gq_long_pop_arm_index]);
+}

inst/stan/lm-stein-fojo/functions.stan

@@ -22,13 +22,5 @@ functions {
         );
         return result;
     }
-    vector lm_predict_value(vector time, matrix long_gq_parameters) {
-        return sld(
-            time,
-            long_gq_parameters[,1],
-            long_gq_parameters[,2],
-            long_gq_parameters[,3]
-        );
-    }
 }
 

inst/stan/lm-stein-fojo/model.stan

@@ -86,17 +86,3 @@ model {
 {%- endif -%}
 }
 
-generated quantities {
-    //
-    // Source - lm-stein-fojo/model.stan
-    //
-    matrix[n_subjects, 3] long_gq_parameters;
-    long_gq_parameters[, 1] = lm_gsf_psi_bsld;
-    long_gq_parameters[, 2] = lm_gsf_psi_ks;
-    long_gq_parameters[, 3] = lm_gsf_psi_kg;
-
-    matrix[gq_n_quant, 3] long_gq_pop_parameters;
-    long_gq_pop_parameters[, 1] = exp(lm_sf_mu_bsld[gq_long_pop_study_index]);
-    long_gq_pop_parameters[, 2] = exp(lm_sf_mu_ks[gq_long_pop_arm_index]);
-    long_gq_pop_parameters[, 3] = exp(lm_sf_mu_kg[gq_long_pop_arm_index]);
-}

inst/stan/lm-stein-fojo/quantities.stan

@@ -0,0 +1,26 @@
+
+functions {
+    vector lm_predict_value(vector time, matrix long_gq_parameters) {
+        return sld(
+            time,
+            long_gq_parameters[,1],
+            long_gq_parameters[,2],
+            long_gq_parameters[,3]
+        );
+    }
+}
+
+generated quantities {
+    //
+    // Source - lm-stein-fojo/quantities.stan
+    //
+    matrix[n_subjects, 3] long_gq_parameters;
+    long_gq_parameters[, 1] = lm_gsf_psi_bsld;
+    long_gq_parameters[, 2] = lm_gsf_psi_ks;
+    long_gq_parameters[, 3] = lm_gsf_psi_kg;
+
+    matrix[gq_n_quant, 3] long_gq_pop_parameters;
+    long_gq_pop_parameters[, 1] = exp(lm_sf_mu_bsld[gq_long_pop_study_index]);
+    long_gq_pop_parameters[, 2] = exp(lm_sf_mu_ks[gq_long_pop_arm_index]);
+    long_gq_pop_parameters[, 3] = exp(lm_sf_mu_kg[gq_long_pop_arm_index]);
+}