Incorrect initial root function values for piecewise-derived events

[dweindl]

For all Piecewise or Boolean expressions we create root functions to stop at the discontinuities.
Currently, their initial value is incorrectly set to true, leading to incorrect simulation results if a piecewise condition is already true at $t_0$.

[dweindl]

The issue might be more nuanced than that.

Here a reproducer:

import amici
from amici.antimony_import import antimony2amici

antimony2amici("""
a = 0
b = 1
a' = 0
ge0 := time >= a
ge1 := time >= b
gt0 := time > a
gt1 := time > b
""", model_name="bla", output_dir="bla")

model_module = amici.import_model_module("bla", "bla")

model = model_module.get_model()
model.setTimepoints([0, 1, 2])

solver = model.getSolver()
rdata = amici.runAmiciSimulation(model, solver)

print(rdata)
print("ts", rdata.ts)
print("ge0", rdata.by_id("ge0"))
print("ge1", rdata.by_id("ge1"))
print("gt0", rdata.by_id("gt0"))
print("gt1", rdata.by_id("gt1"))

ts [0. 1. 2.]
ge0 [1. 1. 1.] # ok
ge1 [0. 1. 1.] # ok
gt0 [0. 0. 0.] # INCORRECT! should be [0. 1. 1.] or at least [1. 1. 1.] until there is a solution to https://github.com/AMICI-dev/AMICI/issues/2707
gt1 [0. 1. 1.] # should be [0. 0. 1.] -> https://github.com/AMICI-dev/AMICI/issues/2707

(The test case added in #2725 can also easily be updated to trigger this issue.)

[FFroehlich]

I thought starting in a root would throw an error. Probably the best way to handle this is start the simulation at 0+eps, but that will make it difficult to make ge0 work at the same time.

In all honesty, I don’t think there is a good way of making this stuff work in general. It might be the better approach to throw informative errors/warnings.

[dweindl]

Thanks. I didn't encounter any error/warning there. I will see if I can find where that might have been supposed to happen.