proteus.DiagUtils module
Module for diagnostic utilities
- proteus.DiagUtils.L2errorFEMvsAF(analyticalFunction, quadraturePointArray, quadratureWeightArray, functionValueArray, T=None)[source]
supposed to be L2 norm of error in vector quantity I think just using dot would cover both scalar and vector case
- proteus.DiagUtils.getQuadraturePhysPointsAndWeights(mesh, femSpace, quadrature, verbose=0)[source]
for convenience, hide steps for generating quadrature points and weights, with Jacobians, on physical mesh based on points and weights on reference element
- returns points array that’s nelem x nquadloc x 3
weight array that’s nelem x nquadloc
- proteus.DiagUtils.getFEMvals(u, xiArray, verbose=0)[source]
for convenience, hide steps for generating finite element solution at physical points corresponding to reference points held in xiArray
returns array that’s nelem x npointloc
- proteus.DiagUtils.testCrRavNodalBasis(nd, verbose=0)[source]
test local Crouzeix-Raviart element space
- proteus.DiagUtils.testQuadNodalBasis(nd, verbose=0)[source]
test local P2 nodal finite element space