diff options
| author | Johannes Ranke <jranke@uni-bremen.de> | 2019-05-02 13:17:05 +0200 |
|---|---|---|
| committer | Johannes Ranke <jranke@uni-bremen.de> | 2019-05-02 13:17:05 +0200 |
| commit | 70591022c07f0e8fb4dd67789b7c8d78af8ebc18 (patch) | |
| tree | acaecfce5ae304cfc48b111c6db24a3f2ed5c83d /docs/articles/web_only/compiled_models.html | |
| parent | 380a29e81f88cd80c9c6915200ddc7054c8a085a (diff) | |
Better initials for error model parameters
- Also make it possible to specify initial values for error model
parameters.
- Run tests
- Rebuild docs
Diffstat (limited to 'docs/articles/web_only/compiled_models.html')
| -rw-r--r-- | docs/articles/web_only/compiled_models.html | 18 |
1 files changed, 9 insertions, 9 deletions
diff --git a/docs/articles/web_only/compiled_models.html b/docs/articles/web_only/compiled_models.html index 597b7c55..ee279f51 100644 --- a/docs/articles/web_only/compiled_models.html +++ b/docs/articles/web_only/compiled_models.html @@ -88,7 +88,7 @@ <h1>Performance benefit by using compiled model definitions in mkin</h1> <h4 class="author">Johannes Ranke</h4> - <h4 class="date">2019-04-10</h4> + <h4 class="date">2019-05-02</h4> <div class="hidden name"><code>compiled_models.Rmd</code></div> @@ -163,14 +163,14 @@ ## Warning in mkinfit(SFO_SFO, FOCUS_2006_D, solution_type = "deSolve", quiet ## = TRUE): Observations with value of zero were removed from the data</code></pre> <pre><code>## test replications elapsed relative user.self sys.self -## 3 deSolve, compiled 3 3.215 1.000 3.213 0 -## 1 deSolve, not compiled 3 42.468 13.209 42.445 0 -## 2 Eigenvalue based 3 4.666 1.451 4.663 0 +## 3 deSolve, compiled 3 3.178 1.000 3.176 0 +## 1 deSolve, not compiled 3 28.700 9.031 28.685 0 +## 2 Eigenvalue based 3 4.401 1.385 4.398 0 ## user.child sys.child ## 3 0 0 ## 1 0 0 ## 2 0 0</code></pre> -<p>We see that using the compiled model is by a factor of around 13 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> +<p>We see that using the compiled model is by a factor of around 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> </div> <div id="model-that-can-not-be-solved-with-eigenvalues" class="section level2"> <h2 class="hasAnchor"> @@ -214,14 +214,14 @@ ## Warning in mkinfit(FOMC_SFO, FOCUS_2006_D, quiet = TRUE): Observations with ## value of zero were removed from the data</code></pre> <pre><code>## test replications elapsed relative user.self sys.self -## 2 deSolve, compiled 3 4.906 1.000 4.902 0 -## 1 deSolve, not compiled 3 70.459 14.362 70.421 0 +## 2 deSolve, compiled 3 4.549 1.000 4.547 0 +## 1 deSolve, not compiled 3 49.752 10.937 49.729 0 ## user.child sys.child ## 2 0 0 ## 1 0 0</code></pre> -<p>Here we get a performance benefit of a factor of 14 using the version of the differential equation model compiled from C code!</p> +<p>Here we get a performance benefit of a factor of 11 using the version of the differential equation model compiled from C code!</p> <p>This vignette was built with mkin 0.9.49.4 on</p> -<pre><code>## R version 3.5.3 (2019-03-11) +<pre><code>## R version 3.6.0 (2019-04-26) ## Platform: x86_64-pc-linux-gnu (64-bit) ## Running under: Debian GNU/Linux 9 (stretch)</code></pre> <pre><code>## CPU model: AMD Ryzen 7 1700 Eight-Core Processor</code></pre> |