Now I understand. Thanks. In my case I¹m calculating three QoI for an
optimization, but I only want one for the adjoint_error_estimator. I also
use the QoISet for those quantities that do not require an adjoint
analysis for the sensitivity analysis.
Miguel
On 2/22/16, 9:53 AM, "Roy Stogner" <royst...@ices.utexas.edu> wrote:
>
>
>On Mon, 22 Feb 2016, Salazar De Troya, Miguel wrote:
>
>> inline
>> bool QoISet::has_index(unsigned int i) const
>> {
>> return (_indices.size() <= i || _indices[i]);
>> }
>>
>> Why does it accept as valid indices greater than _indices.size()
>> (_indices.size() <= i ) ? For instance, say I have three QoI, but I
>> want only to calculate the second QoI; therefore I create a QoISet
>> and call add_index(1). This is going to resize _indices to two and
>> make them all true (_indices.resize(i+1, true)). Now I am
>> calculating 0 and 1, but also when I call QoISet::has_index(2) it is
>> also going to accept it as valid. There must be something I am
>> missing in the way QoISet is used.
>
>You're just missing a questionable design decision, IIRC. The idea
>was that QoISet::has_index(i) would default to true (because most
>people don't bother defining a QoI unless they want to calculate it),
>and so the way to define a more limited QoISet is to remove QoIs from
>the set, not to add them.
>---
>Roy
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
Libmesh-users mailing list
Libmesh-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-users
