-
Notifications
You must be signed in to change notification settings - Fork 108
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Implement ForwardDiff in master solvers and general DiffEq problems on QO types #409
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #409 +/- ##
==========================================
- Coverage 97.03% 96.70% -0.34%
==========================================
Files 18 19 +1
Lines 1554 1701 +147
==========================================
+ Hits 1508 1645 +137
- Misses 46 56 +10 ☔ View full report in Codecov by Sentry. |
@Krastanov Our current ForwardDiff tests are a bit finicky because we are testing numerical equality between ForwardDiff and some implementation of finite difference methods. Should we just increase the abstol or would that be too fragile? Or maybe there's a better way to test ForwardDiff here? Maybe we just use |
increasing the tolerance should be fine (not quite sure whether it is the abs or rel one though) |
@Krastanov is this good to merge? |
The tests currently only test whether operators and their linear algebra works, right? There are no new tests that check the actual solvers? Should we be closing #357 (I guess what tests we need here depends on the answer to that question)? Do Could you add what you have in the first comment (the PR description) to the tests as well? |
@Krastanov the tests do check the master solvers ( |
The
where |
Yes, I can do that. |
Wonderful work! Thank you for contributing it! |
DiffEqBase.promote_u0
methods for QO types and removedrebuild
because it was no longer needed. Once Broadcasting Kets and Operators QuantumOpticsBase.jl#172 is merged, we can use ForwardDiff in the standard SciML approach. For example: