-
In the medium-long term, it makes little sense to continue to support a tape-based AD package when there is excellent work going on elsewhere in the Julia community; see [Capstan](https://github.com/J…
-
Hi, great work!
Any plans to add adjoint sensitivities?
Kr, oyvind?
-
Right now, a `full_tx` is an SoA (“struct of arrays”). This means, at minimum, we include 8 extra bytes in its representation (the count of witnesses is superfluous because all valid transactions must…
-
During early stages of the lingcod work, model input and output files were committed to this repository. However, it became clear that they were adding clutter and focus shifted to sharing models only…
-
Hi All,
I've been trying to get the forward activations and backward sensitivies for layers within my model. The `forward_pre_hook` works as expected but `backward_full_hook` fails to get called.
…
-
(continued from https://github.com/invenia/Nabla.jl/issues/81, cc @willtebbutt)
> This means that we would need a slightly more general interface for linear algebra, and would certainly require dif…
-
I have a STIR parameter file for a reconstruction. At the bottom, outside of the objective function, I set the verbosity:
```
OSMAPOSLParameters :=
objective function type:= PoissonLogLikelihoodW…
-
Hi Ivan, this firedrake extension is awesome. Thanks for putting it together. I am interested in using TSAdjoint. I wonder if you have short-term plans to add it (less than two weeks or so). Otherwise…
-
The operation that I'm trying to do is take a five-component vector whose entries correspond to the degrees of freedom in a 3x3 traceless, symmetric tensor then find a rotation matrix whose columns ar…
-
I get confused about utilizing a Monte Carlo simulation method by conducting "Ecopath_matlab" . I want examine the robustness of the indicators against parameters (B, P/B,Q/B,and EE) uncertainty and …