Solve merge conflicts and merge 906 into main

[digicosmos86]

@AlexanderFengler can you give this a quick approval? When merging PRs this afternoon, I somehow merged them into a separate branch instead of main. This is just to correct the mistake

[AlexanderFengler]

ok @digicosmos86 . Can you just check about the test failure there? That one of the new models, slightly suspicious, but the respective PR for that model had all tests passing....

[digicosmos86]

ok @digicosmos86 . Can you just check about the test failure there? That one of the new models, slightly suspicious, but the respective PR for that model had all tests passing....

Checking now

[digicosmos86]

ok @digicosmos86 . Can you just check about the test failure there? That one of the new models, slightly suspicious, but the respective PR for that model had all tests passing....

@AlexanderFengler Looks like these tests used to be skipped for older SSMS versions. Now that we've updated ssms dependency, these tests are now available:

@pytest.mark.skipif(
    "poisson_race" not in model_config,
    reason="poisson_race not available in installed ssms",
)
class TestPoissonRaceSimulator:

model_config comes from ssms directly

[AlexanderFengler]

ok @digicosmos86 . Can you just check about the test failure there? That one of the new models, slightly suspicious, but the respective PR for that model had all tests passing....

@AlexanderFengler Looks like these tests used to be skipped for older SSMS versions. Now that we've updated ssms dependency, these tests are now available:

@pytest.mark.skipif(
    "poisson_race" not in model_config,
    reason="poisson_race not available in installed ssms",
)
class TestPoissonRaceSimulator:

model_config comes from ssms directly

I see ok. I think this is slightly weird now because we probably were missing an upper bound on ssm-simulators so this is more something to be fixed in a PR dedicated to the actual release? Here, actually let's just put an upper bound on ssm-simulators and then relax that in the release PR and work through any test problems there. Does that make sense @digicosmos86 ?

[digicosmos86]

ok @digicosmos86 . Can you just check about the test failure there? That one of the new models, slightly suspicious, but the respective PR for that model had all tests passing....

@AlexanderFengler Looks like these tests used to be skipped for older SSMS versions. Now that we've updated ssms dependency, these tests are now available:

@pytest.mark.skipif(
    "poisson_race" not in model_config,
    reason="poisson_race not available in installed ssms",
)
class TestPoissonRaceSimulator:

model_config comes from ssms directly

I see ok. I think this is slightly weird now because we probably were missing an upper bound on ssm-simulators so this is more something to be fixed in a PR dedicated to the actual release? Here, actually let's just put an upper bound on ssm-simulators and then relax that in the release PR and work through any test problems there. Does that make sense @digicosmos86 ?

No rush for this. There is another PR coming for choice-only models so I don't think it's realistic for the release to go out this week. I can disable these tests for now in HSSM and then we can re-enable these tests in a future PR. How about that?

[AlexanderFengler]

@digicosmos86 should be good now, we can relegate fixing the poisson stuff to a separate PR, I'll take care of it.

[digicosmos86]

@digicosmos86 should be good now, we can relegate fixing the poisson stuff to a separate PR, I'll take care of it.

Cool thanks @AlexanderFengler! Can you give this an approval?

[AlexanderFengler]

lgtm. Thanks @digicosmos86