-
It's ocassionally not possible to test `@non_differentiable` using `test_rrule` because finite differencing gives the wrong / unreliable results.
For example, this is the case in the `Vector{Floa6…
-
The "pyrely.performance" subpackage should provide the API for performance functions (scalar output only in a first time). The following features need to be studied:
- Should the user provide a classi…
-
Some remarks on
https://github.com/ANU-RSES-Education/EMSC-4033-2023/blob/main/ProjectPlans/ProjectPlan_u7061118/ProjectPlanner.md
- The project proposes to use pseudospectral methods which have…
-
For parallel stencil computations (e.g. finite-difference methods), it would be nice to have partitioned multidimensional arrays, with support for arbitrary thickness "ghost" overlap regions (so that …
-
**Abstract**
Better documentation of the 1D (flame) solver is needed, in a couple of areas.
**Proposed Solution**
_API Documentation_: Many member variables and some member functions of classes u…
speth updated
3 weeks ago
-
### Describe the bug
I try to do the finite difference force test on NiMnTi with nspin=4 noncollinear, and the position of Ni is moved along X axis with 0.01 bohr each step.
As shown in below blue l…
-
Rewrite using Finite differences to see if we recover the true param values.
-
I'm trying to build this with python 3 and matplotlib 2.2.2 but it's getting angry about the Triangulation attributes, it seems like the matplotlib folks updated the Triangulation attributes a bit for…
-
**Is your feature request related to a problem? Please describe.**
I'm working on a project where it's important to estimate higher-order derivatives (e.g. 2nd, 3rd, 4th, and potentially mixed)…
-
In GitLab by @mayani_s on Aug 17, 2022, 15:20
Add a new Finite differences Time domain electromagnetic solver, based on MITHRA: https://github.com/aryafallahi/mithra.