Arun Lakshmanan

Implicit params: no gradient for `Array` element of `Vector{AbstractArray}` when parent `Vector` used in AD

3

I’m seeing incorrectly computed gradients when I use a vector of matrices as a model in Flux. The implicit gradients compute to 0 but the structural gradients seem to be...

Discrepancy between Float32 and Float64

1

MWE: ```julia julia> f(x) = sin(x)^2 f (generic function with 1 method) julia> f(-1..1) [0, 0.708074] julia> f(-1f0..1f0) [-0.708074f0, 0.708074f0] ``` Is this simply a duplicate of #415?

Version Update

9

Keep this issue open to trigger the registrator. @JuliaRegistrator register()