chore(lint/round): build docs, lint files, and round cores output so doesn't change with every run (as it is time taken, which we would expect to vary slightly)

Author: amyheather

DESCRIPTION

@@ -24,7 +24,9 @@ Imports:
     rlang,
     tidyr,
     R6,
-    tidyselect
+    tidyselect,
+    future,
+    future.apply
 Suggests:
     testthat (>= 3.0.0),
     lintr,

R/defaults.R

@@ -93,7 +93,7 @@ Defaults <- R6Class( # nolint
 #' the class without needing to run `Defaults[["new"]]()` every time.
 #'
 #' @return Instance of the Defaults class.
-#' 
+#'
 #' @export
 #' @rdname defaults
 

R/model.R

@@ -28,7 +28,7 @@ model <- function(run_number, param_class, set_seed = TRUE) {
   valid_inputs(run_number, param)
 
   # Set random seed based on run number
-  if (set_seed == TRUE) {
+  if (set_seed) {
     set.seed(run_number)
   }
 

R/trial.R

@@ -30,9 +30,12 @@ trial <- function(param_class) {
   # within the function, and we don't want to override future.seed
   results <- future_lapply(
     1L:n_runs,
-    function(i) simulation::model(
-      run_number = i, param_class = param_class, set_seed = FALSE),
-    future.seed = 123456
+    function(i) {
+      simulation::model(run_number = i,
+                        param_class = param_class,
+                        set_seed = FALSE)
+    },
+    future.seed = 123456L
   )
 
   # Combine the results from multiple replications into just two dataframes

man/model.Rd

@@ -45,12 +45,11 @@ devtools::install()
     ##    checking DESCRIPTION meta-information ...  ✔  checking DESCRIPTION meta-information
     ##   ─  checking for LF line-endings in source and make files and shell scripts
     ## ─  checking for empty or unneeded directories
-    ##      Removed empty directory ‘simulation/tests/testthat/_snaps’
-    ##    Omitted ‘LazyData’ from DESCRIPTION
+    ##      Omitted ‘LazyData’ from DESCRIPTION
     ##   ─  building ‘simulation_0.1.0.tar.gz’
     ##      
     ## Running /opt/R/4.4.1/lib/R/bin/R CMD INSTALL \
-    ##   /tmp/RtmpRZdNqo/simulation_0.1.0.tar.gz --install-tests 
+    ##   /tmp/Rtmpbqk15J/simulation_0.1.0.tar.gz --install-tests 
     ## * installing to library ‘/home/amy/.cache/R/renv/library/rap_template_r_des-cd7d6844/linux-ubuntu-noble/R-4.4/x86_64-pc-linux-gnu’
     ## * installing *source* package ‘simulation’ ...
     ## ** using staged installation
@@ -423,7 +422,7 @@ print(table_latex)
 ```
 
     ## % latex table generated in R 4.4.1 by xtable 1.8-4 package
-    ## % Mon Jan 27 15:52:00 2025
+    ## % Mon Jan 27 16:03:02 2025
     ## \begin{table}[ht]
     ## \centering
     ## \begin{tabular}{rrllll}
@@ -532,7 +531,7 @@ print(sensitivity_table_latex)
 ```
 
     ## % latex table generated in R 4.4.1 by xtable 1.8-4 package
-    ## % Mon Jan 27 15:52:05 2025
+    ## % Mon Jan 27 16:03:07 2025
     ## \begin{table}[ht]
     ## \centering
     ## \begin{tabular}{rrl}
@@ -574,12 +573,12 @@ tail(result[["arrivals"]])
 ```
 
     ##           name start_time end_time activity_time resource replication
-    ## 160 patient104   51.17679       NA            NA    nurse           0
-    ## 161 patient110   53.57293       NA            NA    nurse           0
-    ## 162 patient148   74.24750       NA            NA    nurse           0
-    ## 163 patient149   74.50965       NA            NA    nurse           0
+    ## 160  patient97   47.84534       NA            NA    nurse           0
+    ## 161 patient155   77.41953       NA            NA    nurse           0
+    ## 162  patient66   35.72669       NA            NA    nurse           0
+    ## 163  patient52   27.21852       NA            NA    nurse           0
     ## 164 patient156   77.59569       NA            NA    nurse           0
-    ## 165 patient158   78.34954       NA            NA    nurse           0
+    ## 165 patient160   78.41585       NA            NA    nurse           0
 
 ## Example run with logs
 
@@ -818,8 +817,8 @@ verbose_run[["arrivals"]]
     ## 22 patient12 46.4500010 86.039329    39.5893285    nurse           0
     ## 23 patient22 77.2295939 87.175152     9.9455579    nurse           0
     ## 24 patient25 86.1477594 91.726053     5.5782935    nurse           0
-    ## 25 patient24 84.4587897        NA            NA    nurse           0
-    ## 26 patient26 99.0189155        NA            NA    nurse           0
+    ## 25 patient26 99.0189155        NA            NA    nurse           0
+    ## 26 patient24 84.4587897        NA            NA    nurse           0
     ## 27 patient23 82.4020925        NA            NA    nurse           0
 
 ## Calculate run time
@@ -835,4 +834,4 @@ seconds <- as.integer(runtime %% 60L)
 print(sprintf("Notebook run time: %dm %ds", minutes, seconds))
 ```
 
-    ## [1] "Notebook run time: 0m 18s"
+    ## [1] "Notebook run time: 0m 17s"

man/trial.Rd

@@ -145,7 +145,7 @@ confidence_interval_method <- function(replications, desired_precision, metric,
     warning("Running ", replications, " replications did not reach ",
             "desired precision (", desired_precision, ").")
   }
-  
+
   # Plot the cumulative mean and confidence interval
   p <- ggplot(cumulative, aes(x = .data[["replications"]],
                               y = .data[["cumulative_mean"]])) +
@@ -165,7 +165,7 @@ confidence_interval_method <- function(replications, desired_precision, metric,
 
   # Save the plot
   ggsave(filename = path, width = 6.5, height = 4L, bg = "white")
-  
+
   return(cumulative)
 }
 ```
@@ -187,8 +187,8 @@ ci_df <- confidence_interval_method(
 
 # View first ten rows were percentage deviation is below 5
 ci_df %>%
-  filter(perc_deviation < 5) %>%
-  head(10)
+  filter(perc_deviation < 5L) %>%
+  head(10L)
 
 # View plot
 include_graphics(path)
@@ -211,8 +211,8 @@ ci_df <- confidence_interval_method(
 
 # View first ten rows were percentage deviation is below 5
 ci_df %>%
-  filter(perc_deviation < 5) %>%
-  head(10)
+  filter(perc_deviation < 5L) %>%
+  head(10L)
 
 # View plot
 include_graphics(path)
@@ -243,8 +243,9 @@ run_cores <- function(n_cores, file, model_param = NULL) {
     param_class[["update"]](list(cores = i))
     invisible(trial(param_class))
 
-    # Record time taken
-    cores_time <- as.numeric(Sys.time() - cores_start, units = "secs")
+    # Record time taken, rounded to nearest .5 dp by running round(x*2)/2
+    cores_time <- round(
+      as.numeric(Sys.time() - cores_start, units = "secs")*2)/2
     speed[[i]] <- list(cores = i, run_time = round(cores_time, 3L))
   }
 
@@ -254,7 +255,7 @@ run_cores <- function(n_cores, file, model_param = NULL) {
   # Generate plot
   p <- ggplot(speed_df, aes(x = .data[["cores"]], y = .data[["run_time"]])) +
     geom_line() +
-    labs(x = "Cores", y = "Run time (seconds)") +
+    labs(x = "Cores", y = "Run time (rounded to nearest .5 seconds)") +
     theme_minimal()
 
   # Save plot
@@ -270,7 +271,7 @@ run_cores <- function(n_cores, file, model_param = NULL) {
 Setting up and managing a parallel cluster takes extra time. For small tasks or few iterations, this extra time can be more than the time saved by running in parallel.
 
 ```{r}
-run_cores(5, "cores1.png")
+run_cores(5L, "cores1.png")
 ```
 
 Having increased the simulation length, we now see that parallelisation is decreasing the model run time.
@@ -280,7 +281,7 @@ However, when you use more cores, the data needs to be divided and sent to more
 The optimal number of cores will vary depending on your model parameters and machine.
 
 ```{r}
-run_cores(5, "cores2.png", list(data_collection_period = 100000L))
+run_cores(5L, "cores2.png", list(data_collection_period = 100000L))
 ```
 
 ## Run time