We should consider a basic set of tests that run each time a piece of code is checked in. This would also be run against each pull request to ensure that broken code is not merged into the master branch.
At a minimum, this could just check that the code compiles and runs the initial conditions and maybe a very simple test case. Additionally, we could consider adding a set of benchmarking data to compare against and raise an error if there are differences relative to this new data set.
We should consider a basic set of tests that run each time a piece of code is checked in. This would also be run against each pull request to ensure that broken code is not merged into the master branch.
At a minimum, this could just check that the code compiles and runs the initial conditions and maybe a very simple test case. Additionally, we could consider adding a set of benchmarking data to compare against and raise an error if there are differences relative to this new data set.