Integrating Observables
Features
- "Fixed" or "Adaptive" mode
- In fixed mode there are a fixed set of energy/momentum points; no extra re-computation will be done
- In adaptive mode extra computations will be launched on-fly. This is guaranteed to work, but may take up to 2x longer than using fixed mode
- Adaptive mode leverages the
cquad
algorithm. This uses Clenshaw-Curtis, but in principle we could give up the "doubly adaptive" part of the algorithm and use Gauss-Kronrod or some other rule. For the adaptive to work we just need a rule with an error estimator. - In fixed mode we can choose any scheme we wish, but it is a good idea to choose one with an error estimator, otherwise it is impossible to know how accurate your results are