aboutsummaryrefslogtreecommitdiff
path: root/vignettes/compiled_models.html
diff options
context:
space:
mode:
authorJohannes Ranke <jranke@uni-bremen.de>2015-06-23 12:24:10 +0200
committerJohannes Ranke <jranke@uni-bremen.de>2015-06-23 12:24:10 +0200
commitb2aa7f7d6180edda1ac1ba4d8c3c3d364cbd3754 (patch)
treed8e97dac4fc8941fb02784cf4fe3ca6479c09d55 /vignettes/compiled_models.html
parent9aea634a324d567a2d99f22b5c6446c0afb2c83c (diff)
Vignettes rebuilt by staticdocs::build_site() for static documentation on r-forge
Diffstat (limited to 'vignettes/compiled_models.html')
-rw-r--r--vignettes/compiled_models.html30
1 files changed, 15 insertions, 15 deletions
diff --git a/vignettes/compiled_models.html b/vignettes/compiled_models.html
index 814f3a52..2fbef8ae 100644
--- a/vignettes/compiled_models.html
+++ b/vignettes/compiled_models.html
@@ -99,18 +99,18 @@ smb.1 &lt;- summary(mb.1)[-1]
rownames(smb.1) &lt;- c(&quot;deSolve, not compiled&quot;, &quot;Eigenvalue based&quot;, &quot;deSolve, compiled&quot;)
print(smb.1)</code></pre>
<pre><code>## min lq mean median uq
-## deSolve, not compiled 6650.2684 6684.4530 6774.1607 6718.6377 6836.1068
-## Eigenvalue based 903.5520 916.8598 927.3873 930.1676 939.3049
-## deSolve, compiled 751.1205 752.5239 756.1227 753.9273 758.6238
+## deSolve, not compiled 6760.8015 6766.0464 6785.8509 6771.2913 6798.3756
+## Eigenvalue based 930.9980 944.5895 973.5743 958.1810 994.8625
+## deSolve, compiled 758.2721 811.0584 836.3348 863.8448 875.3661
## max neval
-## deSolve, not compiled 6953.5760 3
-## Eigenvalue based 948.4423 3
-## deSolve, compiled 763.3202 3</code></pre>
-<p>We see that using the compiled model is by a factor of 8.9 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>
+## deSolve, not compiled 6825.4600 3
+## Eigenvalue based 1031.5439 3
+## deSolve, compiled 886.8875 3</code></pre>
+<p>We see that using the compiled model is by a factor of 7.8 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[&quot;median&quot;]/smb.1[&quot;deSolve, compiled&quot;, &quot;median&quot;]</code></pre>
<pre><code>## median
-## deSolve, not compiled 8.911519
-## Eigenvalue based 1.233763
+## deSolve, not compiled 7.838551
+## Eigenvalue based 1.109205
## deSolve, compiled 1.000000</code></pre>
</div>
<div id="benchmark-for-a-model-that-can-not-be-solved-with-eigenvalues" class="section level2">
@@ -127,15 +127,15 @@ print(smb.1)</code></pre>
smb.2 &lt;- summary(mb.2)[-1]
rownames(smb.2) &lt;- c(&quot;deSolve, not compiled&quot;, &quot;deSolve, compiled&quot;)
print(smb.2)</code></pre>
-<pre><code>## min lq mean median uq
-## deSolve, not compiled 14.32061 14.336413 14.380847 14.352216 14.410966
-## deSolve, compiled 1.34366 1.344778 1.371116 1.345897 1.384844
+<pre><code>## min lq mean median uq
+## deSolve, not compiled 14.231863 14.287058 14.336780 14.342252 14.389238
+## deSolve, compiled 1.344761 1.345452 1.361713 1.346142 1.370189
## max neval
-## deSolve, not compiled 14.469716 3
-## deSolve, compiled 1.423791 3</code></pre>
+## deSolve, not compiled 14.436224 3
+## deSolve, compiled 1.394235 3</code></pre>
<pre class="r"><code>smb.2[&quot;median&quot;]/smb.2[&quot;deSolve, compiled&quot;, &quot;median&quot;]</code></pre>
<pre><code>## median
-## deSolve, not compiled 10.66368
+## deSolve, not compiled 10.65434
## deSolve, compiled 1.00000</code></pre>
<p>Here we get a performance benefit of a factor of 10.7 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>

Contact - Imprint