The main aim of this PR is to get go_mesh_seq.py under the test harness so that we can tackle #55. It goes as far as covering all of the code except the bit that issue addresses. In a future PR, we will introduce a broken test for that part and then fix it.
This PR also moves indicators2estimator from being a function defined in error_estimation.py to being a method of GoalOrientedMeshSeq. This felt natural because it only has a single call site, in that class' fixed_point_iteration method.
The global_enrichment method of GoalOrientedMeshSeq is dropped as it is unused.
Closes #68. Partially addresses #42.
The main aim of this PR is to get
go_mesh_seq.pyunder the test harness so that we can tackle #55. It goes as far as covering all of the code except the bit that issue addresses. In a future PR, we will introduce a broken test for that part and then fix it.This PR also moves
indicators2estimatorfrom being a function defined inerror_estimation.pyto being a method ofGoalOrientedMeshSeq. This felt natural because it only has a single call site, in that class'fixed_point_iterationmethod.The
global_enrichmentmethod ofGoalOrientedMeshSeqis dropped as it is unused.