Christopher Tessum

It's the ` %4 = call double @j_sin_3353(double %2)`, right?

Still there: ```julia @code_llvm optimize=true prob.f([0.0, 0.0], 0.0, 0.0) ``` ```julia @ //.julia/packages/SciMLBase/hq1ku/src/scimlfunctions.jl:2299 within `ODEFunction` define nonnull {}* @julia_ODEFunction_2895({ [1 x i8], { { i64, {}*, { {}*, i8, {}*,...

This is fixed in mtk v10. Thanks!

We're just trying to understand how it works. Would a reasonable summary be that if if there's a single system of equations that exactly describes the data (like if we're...

The next question would be if we want to a single implementation to give us all of the different plausible explanations, how would we do that? Just run it multiple...

Thanks! I didn't realize that the constants could be overridden, so that could be a good short term fix, but obviously that wouldn't be practical for a large scale model....

cc @jialinl6

I think the issue is that MArrays are sometimes allocated to the heap and sometimes allocated to the stack: https://stackoverflow.com/questions/74793823/mutable-data-types-that-use-stack-allocation So the allocation tests seem to pass in some cases...

cc @jialinl6

Thanks. I can confirm that the method in the previous doc worked with the previous version of the library, but it doesn't seem to work with v1.0.0.