diff options
Diffstat (limited to 'vignettes/compiled_models.html')
| -rw-r--r-- | vignettes/compiled_models.html | 34 |
1 files changed, 17 insertions, 17 deletions
diff --git a/vignettes/compiled_models.html b/vignettes/compiled_models.html index ed61b47a..0b77f1c2 100644 --- a/vignettes/compiled_models.html +++ b/vignettes/compiled_models.html @@ -98,19 +98,19 @@ mb.1 <- microbenchmark( smb.1 <- summary(mb.1)[-1] rownames(smb.1) <- c("deSolve, not compiled", "Eigenvalue based", "deSolve, compiled") print(smb.1)</code></pre> -<pre><code>## min lq mean median uq -## deSolve, not compiled 6585.7039 6651.4937 6685.6248 6717.2836 6735.5853 -## Eigenvalue based 971.2893 981.5618 998.2746 991.8344 1011.7673 -## deSolve, compiled 760.5522 765.4274 780.3243 770.3026 790.2103 -## max neval -## deSolve, not compiled 6753.8871 3 -## Eigenvalue based 1031.7003 3 -## deSolve, compiled 810.1179 3</code></pre> -<p>We see that using the compiled model is by a factor of 8.7 faster than using the R version with the default ode solver, and it is even faster than the Eigenvalue based solution implemented in R which does not need iterative solution of the ODEs:</p> +<pre><code>## min lq mean median uq +## deSolve, not compiled 6737.589 6818.2149 6911.3916 6898.8407 6998.2929 +## Eigenvalue based 945.433 968.8592 979.7477 992.2854 996.9051 +## deSolve, compiled 744.785 748.8107 770.7521 752.8364 783.7357 +## max neval +## deSolve, not compiled 7097.745 3 +## Eigenvalue based 1001.525 3 +## deSolve, compiled 814.635 3</code></pre> +<p>We see that using the compiled model is by a factor of 9.2 faster than using the R version with the default ode solver, and it is even faster than the Eigenvalue based solution implemented in R which does not need iterative solution of the ODEs:</p> <pre class="r"><code>smb.1["median"]/smb.1["deSolve, compiled", "median"]</code></pre> <pre><code>## median -## deSolve, not compiled 8.720318 -## Eigenvalue based 1.287591 +## deSolve, not compiled 9.163798 +## Eigenvalue based 1.318062 ## deSolve, compiled 1.000000</code></pre> </div> <div id="benchmark-for-a-model-that-can-not-be-solved-with-eigenvalues" class="section level2"> @@ -128,15 +128,15 @@ smb.2 <- summary(mb.2)[-1] rownames(smb.2) <- c("deSolve, not compiled", "deSolve, compiled") print(smb.2)</code></pre> <pre><code>## min lq mean median uq -## deSolve, not compiled 14.271472 14.285039 14.303450 14.298607 14.319440 -## deSolve, compiled 1.350642 1.390549 1.412823 1.430456 1.443914 +## deSolve, not compiled 13.955273 13.961009 14.041563 13.966745 14.084708 +## deSolve, compiled 1.350567 1.371225 1.381397 1.391882 1.396812 ## max neval -## deSolve, not compiled 14.340272 3 -## deSolve, compiled 1.457372 3</code></pre> +## deSolve, not compiled 14.202672 3 +## deSolve, compiled 1.401743 3</code></pre> <pre class="r"><code>smb.2["median"]/smb.2["deSolve, compiled", "median"]</code></pre> <pre><code>## median -## deSolve, not compiled 9.995841 -## deSolve, compiled 1.000000</code></pre> +## deSolve, not compiled 10.03443 +## deSolve, compiled 1.00000</code></pre> <p>Here we get a performance benefit of a factor of 10 using the version of the differential equation model compiled from C code using the inline package!</p> <p>This vignette was built with mkin 0.9.37 on</p> <pre><code>## R version 3.2.1 (2015-06-18) |