-
Currently we choose `firststep = ivp.tmax - ivp.t0`. This seems to work well, but I'm wondering if we could use a better default here. My intuition for example is that we might prefer starting small a…
-
Hi devs!
This repo looks cool :sunglasses:
When I try to calculate finite difference coefficients, I encountered exception `Cannot compute accuracy'. https://github.com/maroba/findiff/blob/e8ca337…
ajz34 updated
3 years ago
-
**Submitting author:** @DrDaveShaw (David Shaw)
**Repository:** https://github.com/DrDaveShaw/INCHEM-Py
**Version:** v1.1
**Editor:** @dhhagan
**Reviewer:** @goldmanm, @khinsen
**Archive:** 10.5281/ze…
-
I looked through the package and it seems like things are already in CuArrays then
https://github.com/climate-machine/Oceananigans.jl/blob/master/src/time_steppers.jl#L42-L78
this should quite r…
-
What is the correct use of the `ulp` method?
For symbolic "exact number" classes I would like to return `0`. Is this correct?
In the following case I'm using an instanceof of the "not exact number…
-
The step size controller which is implemented right now does not support relative tolerances. Ideally we would like to have a controller which closely corresponds to the proportional control algorithm…
-
We came across your software searching for ways to investigate the propagation of traffic waves.
The model needs to run offline, coupled to Python to interface AI toolkits.
Of course we would like t…
-
### How would this new feature help [increase access](https://github.com/processing/p5.js/blob/main/contributor_docs/access.md) to p5.js?
I think it would be nice to have a function that takes a …
-
Hi, I am a ML researcher and majoring in sequential VAE model.
The work ODE2VAE interests me a lot. But I have a question about the ELBO loss function in Equ(16) which is unelated to the code.
![im…
-
The following system of ODEs with variable coefficients gives the wrong answer. In particular, it is the case for when `elif r['d'] == -r['a']:` that the error occurs:
```
eqs = [
Eq(f(x).diff(x), …