Closed AndreyAPopov closed 2 years ago
So here is how all this works in matlab:
>> model = otp.brusselator.presets.Canonical;
>> firstCompRHS = model.RHS(1, :);
>> secondCompRHS = model.RHS(2, :);
>> reverseRHS = [secondCompRHS; firstCompRHS];
>> model.RHS.F(0, model.Y0)
ans =
-6
5
>> reverseRHS.F(0, model.Y0)
ans =
5
-6
This should be ready to merge after some testing
I feel like what we need now is to setup some pretty serious edge-case tests to exhaustively test every possible scenario
As it is now, there is effectively no guarantees that this will work as intended
I say we go ahead and merge this. Clearly it needs additional testing, but proper testing for OTP will be a large undertaking in a separate PR.
I say we go ahead and merge this. Clearly it needs additional testing, but proper testing for OTP will be a large undertaking in a separate PR.
Fine. It is in a usable state. We will get more mileage out of testing by people actually using it.
This is a test as of now.