-
_From @rtrangucci on January 21, 2015 1:15_
the test_gradients test in test_fixture_distr.hpp, test_fixture_cdf.hpp, test_fixture_cdf_log.hpp and test_fixture_ccdf_log.hpp is broken.
The test_gradi…
-
rlouf updated
2 months ago
-
While working at https://github.com/FluxML/NNlib.jl/pull/260 I hit a bug on Zygote master that I managed to reduce to
the following
```julia
julia> f(x) = reshape(x, fill(2, 2)...)
f (generic f…
-
Fix the bug causing the gradients to be off when using zipper meshes.
-
There are basically two reasons to implement rules:
1. to define AD. For example, you do have to tell an AD system _somewhere_ how to differentiate addition and multiplication of floats,
2. to make …
-
Requested by Avraham Adler on stan-users:
Firstly, as a medium-term lurker but first-time poster, I would like to thank the entire Stan development team for the creation and maintenance of Stan.
I a…
-
To-do list so far:
Brownian motions:
- Switch to interval interface.
- [x] Brownian
- [x] BrownianInterval
- [x] BrownianPath (both Python and C++)
- [x] BrownianTree (both Pyt…
-
Are integer types supported? I only get the expected answer for floats or doubles.
Correct answer with type 'float':
```c
#include
#include
#define DTYPE float
extern DTYPE __enzyme_aut…
-
I played with Enzyme a little bit, and I suspect it's not ready for use with our package. It can't differentiate simple ODEs at present. There's a fair amount of linear algebra in OrdinaryDiffEq, so i…
-
```julia
using Distances
using Flux
using Flux: gradient
Xp = randn(500)
X = randn(500)
lossobj = CosineDist()
lossobj(X,Xp)
loss = Xp -> evaluate(lossobj, X,Xp)
gradient(loss, Xp)
```
resu…