When there is higher test coverage of param, and when we have performance tests, would be worthwhile investigating whether cythonizing param gives any useful performance benefits, and if so under what conditions (presumably not for typical param usage).
Might also be useful to cythonize param as part of testing (e.g. can be useful to catch certain classes of error/bad practice).
See: #166
When there is higher test coverage of param, and when we have performance tests, would be worthwhile investigating whether cythonizing param gives any useful performance benefits, and if so under what conditions (presumably not for typical param usage).
Might also be useful to cythonize param as part of testing (e.g. can be useful to catch certain classes of error/bad practice).
See: #166