June 2008 - Comp-phys-alps-users

obs_value_traits< std::valarray<double> >::convergence_type has no == defined
by Johannes Wilms 14 Jun '08

14 Jun '08

I am trying to check a vector observable for convergence by testing its evaluator like: myvectorobseval.converged_errors()==alps::CONVERGED This code works fine for scalar observable evaluators, but for vector observable evaluators I get the following error: error: no match for 'operator==' in 'alps::SimpleObservableEvaluator<T>::converged_errors() const [with T = std::valarray<double>]() == NOT_CONVERGED' Somehow "==" seems to be undefined for obs_value_traits< std::valarray<double> >::convergence_type, so I would be grateful for any advice how to work around this rather strange behaviour. Thank you very much in advance, Johannes

2 1

alea: functions of vector observables
by Johannes Wilms 14 Jun '08

14 Jun '08

Hi everyone, am I right in assuming that the alea library does not yet implement functions of RealVectorObsevaluators that involve scalars or scalar RealObsevaluators? I keep getting errors like: error: no match for 'operator*' in '2.0e+0 * myvectorobseval I would be very grateful if someone could point out a way to make these kinds of functions work, or tell me if this is a planned feature, or, as a kind of last resort, give a hint about how I could possibly implement this. Thank you very much in advance, best regards, Johannes Wilms

2 3

ALPS release 1.3.2
by Matthias Troyer 06 Jun '08

06 Jun '08

Dear colleagues, Following the Sherbrooke/Jouvence summer school we have released a new version of ALPS with several improvements and bug fixes including * a printgraph tool * a bug fix in the fermion model * a new coupled ladder lattice * support for random couplings in disordered models Matthias Troyer

1 0