From a01fe8923ff50dcc76c34fb09ad500f5f5433063 Mon Sep 17 00:00:00 2001
From: Johannes Ranke Johannes
Ranke
- 2023-04-16
+ 2023-05-19
Source: vignettes/web_only/compiled_models.rmd
compiled_models.rmd
## test replications relative elapsed
-## 4 analytical 1 1.000 0.105
-## 3 deSolve, compiled 1 1.276 0.134
-## 2 Eigenvalue based 1 1.762 0.185
-## 1 deSolve, not compiled 1 21.914 2.301
+## 4 analytical 1 1.000 0.108
+## 3 deSolve, compiled 1 1.259 0.136
+## 2 Eigenvalue based 1 1.704 0.184
+## 1 deSolve, not compiled 1 21.574 2.330
We see that using the compiled model is by more than a factor of 10 faster than using deSolve without compiled code.
@@ -248,12 +248,12 @@ compiled code is available. }## Temporary DLL for differentials generated and loaded
## test replications relative elapsed
-## 2 deSolve, compiled 1 1.000 0.176
-## 1 deSolve, not compiled 1 23.938 4.213
+## 2 deSolve, compiled 1 1.000 0.178
+## 1 deSolve, not compiled 1 23.708 4.220
Here we get a performance benefit of a factor of 24 using the version of the differential equation model compiled from C code!
-This vignette was built with mkin 1.2.3 on
-## R version 4.2.3 (2023-03-15)
+This vignette was built with mkin 1.2.5 on
+## R version 4.3.0 (2023-04-21)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Debian GNU/Linux 12 (bookworm)
## CPU model: AMD Ryzen 9 7950X 16-Core Processor
--
cgit v1.2.1