Alex Robson
Alex Robson
Probably this generalises to others in Base.Iterators Sometimes this works in Zygote and sometimes it doesn't - i haven't worked out what the difference is: ``` x, y = rand(3,),...
In FiniteDifferences, this PR allowed custom constructors to be handled as part of `to_vec`: https://github.com/JuliaDiff/FiniteDifferences.jl/pull/181 [here](https://github.com/JuliaDiff/FiniteDifferences.jl/blob/master/src/to_vec.jl#L53) However iiuc `construct` in CRC [here](https://github.com/JuliaDiff/ChainRulesCore.jl/blob/master/src/differentials/composite.jl#L211) doesn't handle this. In this case however, the...
``` R = rand(10,10) A = KeyedArray(R, iter=1:10, iter2=1:10) svd(A).U isa NamedDimsArray ``` Possibly related: https://github.com/mcabbott/AxisKeys.jl/issues/31
```julia using Zygote using LinearAlgebra r = rand(8,8); Σ = r' * r foo(X) = tr(svd(X).U) _orthogonal(X) = svd(X).U * svd(X).V' Zygote.gradient(foo, Σ) # Works Zygote.gradient(foo, 1.0 * Matrix(I(5))) #...
Solves #1607 Similar to MvNormal, extends * + and - to the GenericMvTDist. This will allow for transformations of the form: ``` [1,2] + Diagonal([2,2]) * GenericMvTDist(4.5, zeros(2,), PDMat(I(2))) Distributions.MvTDist(...
In the univariate case we can do: ``` julia> μ + TDist(ν)*σ LocationScale{Float64, Distributions.Continuous, TDist{Float64}}( μ: 1.0 σ: 5.0 ρ: TDist{Float64}(ν=4.5) ) ``` However, we can't do this in the...
I have a pipeline that is producing quite spammed log out put when it repeatedly hits a `@warn`. However, this line does have a `maxlog=1` directive with it. When using...
Having just switched to experimenting with Gadfly, I was confused by the apparent behaviour in decreasing opacity of the plots as the number of points became more dense. Essentially, this...
Minor thing, but this was playing around with the two sets of introductory code: 1) https://github.com/SciML/NeuralPDE.jl/blob/master/README.md#example-solving-2d-poisson-equation-via-physics-informed-neural-networks 2) https://docs.sciml.ai/NeuralPDE/stable/tutorials/gpu/ Essentially, naively extending the example [1] in the readme with the push...
Fixes #https://github.com/invenia/Intervals.jl/issues/159 Plots changes the way it handles line segments. Instead of `NaN` separated, it uses a vector of points (see issue). Should we keep support for Plots < v1.11