-
When we want to draw the graph of the solution of a second-order ODE with u, u' and u'' we need to use a very counterintuitive and redundant syntax.
```julia
f2(du, u, p, t) = -0.1du - 0.8u
u₀ = …
-
Invenio legacy supported queries such as: `citedby:author:ellis`. We need `invenio-search` to be able to support them, too, as this is critical for services such as INSPIRE for users to be able to bro…
-
Hi Zulko,
firstly, thank you for writing ddeint!. I am trying to solve an initial value problem for the case that x''(t) is known and depends on x(t-dt). In scipy without the delay I do this as fol…
-
ACK the numbers are different when I do something completely random, please tell me why !?
-
-
Hi
I need to compute the approximate hessian for a decoder network. The decoder consists of conv2d and upsample layers. Currently, backpack does not supports nn.Upsample. Since it is a non-parametr…
-
## 🐛 Bug
`torch.vmap` could be a great tool to compute only certain parts of higher-order gradients, e.g., the Hessian.
A common use case would be to compute the Laplacian where only the diagonal of…
n-gao updated
3 years ago
-
Zygote v0.6.41, Julia 1.7.3
MWE:
```julia
using Zygote
α, β = randn(2, 2), randn(2, 2)
g(v) = map(eachcol(v), eachcol(β)) do x, y
sum(x.*x.*y)
end |> sum
# this fails
…
-
I try to implement getting the second order gradient by tf_agent.
The reason why I do second order gradient is came from meta-learning algorithm [MAML](https://arxiv.org/abs/1703.03400).
First I c…
-
I need the normal distribution to support second order gradients of Gaussian distribution, since the current implementation relies on broadcasting, which breaks down.
I have quickly hack it as fol…