-
Right now, I'm working on in-place, CPU versions of `kronsum(A,B) = kron(A, oneunit(B)) + kron(oneunit(A), B)`
So far, this works for dense arrays:
```julia
function kronsum!(C::AbstractMatrix,…
-
#22760 added support for `__matmul__` in the coercion model.
We should start using it.
First step: Review the semantics of this operator in major Python software for matrix and tensor computation …
-
How can we represent variable length arrays? Another sort of related question is how to arrays of mixed data types, ala structs/tuples (not unions of types)?
These could come in a couple of forms:
…
-
It's sometimes useful to have multiple layers of blocks, see conversation in
https://github.com/JuliaDiffEq/MultiScaleArrays.jl/issues/25
I think one way to do this would be to support some sor…
-
http://arxiv.org/abs/1503.05671
Any chance someone could implement this for lasagne.updates?
It could be very good as it overcomes some of the issues with HF while also being faster to converge than…
-
I have the example code below, expecting the output to be 0, 2, 5, and 7 with equal probability. But the output from the code below only produce 0 and 2. Did I do somthing wrong on the `CreateQuantumI…
ranvd updated
3 weeks ago
-
Hi,
I am trying to obtain a Kronecker-factored Laplace approximation for an entire Resnet (not only for the last layer) but obtain an error that some layers are not supported (I suspect this is due…
-
I am trying to fit a model in which each sample has three float features (X) and a corresponding set of 13 vectors of 100 points (y).
`Dataset shape (X, y): torch.Size([1500, 3]), torch.Size([1500…
-
This issue is to keep track of making improvements to gcd calculations with sparse polynomials. Currently sparse polynomial gcd is slow particularly when there are many generators. You can read more a…
-
Submitting Author Name: Weicheng Sun
Submitting Author Github Handle: @WeichengSun
Other Package Authors Github handles: (comma separated, delete if none)
Repository: https://github.com/Weich…