search

SciML / JumpProcesses.jl

Build and simulate jump equations like Gillespie simulations and jump diffusions with constant and state-dependent rates and mix with differential equations and scientific machine learning (SciML)
https://docs.sciml.ai/JumpProcesses/stable/
Other
135 stars 35 forks source link

use SII.finalize_parameters_hook! when updating problems #409

Closed isaacsas closed 3 months ago

isaacsas commented 3 months ago

Finishes https://github.com/SciML/JumpProcesses.jl/pull/408

coveralls commented 3 months ago

Pull Request Test Coverage Report for Build 8440125933

Details

Changes Missing Coverage Covered Lines Changed/Added Lines %
src/problem.jl 2 3 66.67%
<!-- Total: 2 3 66.67% -->
Totals Coverage Status
Change from base Build 8422543521: -0.005%
Covered Lines: 2208
Relevant Lines: 2567
💛 - Coveralls
isaacsas commented 3 months ago

If tests pass I'll plan to merge and make a new release.

ChrisRackauckas commented 3 months ago

Awesome, good to see this is finally done.

So then, Catalyst is close?

Sorry for all of the chaos but I think we're getting to stable again.

isaacsas commented 3 months ago

We're getting there! On the jump side all that is left after this is updating the parameter mapper for JumpSystems in MTK to work with SII correctly (i.e. the stuff that handles recalculating mass action rates when parameters change from the symbolic rate constant expressions). I'll then add some more comprehensive tests in MTK to go with that and make sure symbolic indexing works in the various desired ways with jump systems.

I think aside from that @TorkelE has successfully run most Catalyst stuff now with the merged and open PRs we are finalizing. He has really put in a heroic effort updating stuff and patiently addressing my many comments, including a pretty through update of our tests. We are tracking what needs to be done for a MTK9 compatible release in https://github.com/SciML/Catalyst.jl/issues/775 -- we have a bunch of other changes that will go in via other open PRs, but should hopefully get a release in the not too distant future (like getting 1:1 equivalence between the DSL and symbolic interface finally).