What is the best machine learning framework in Julia that works with Metal.jl?

Specific Domains Machine Learning
1

What is the best machine learning framework in Julia that works with Metal.jl?

2

I’m not familiar with Metal, but most ML pkgs in Julia that support arrays support any AbstractArray, and so I expect should work with the MtlArray objects as a special case. Perhaps try a simple linear model from GLM.jl or Flux.jl to start with. For ML tools that consume tables (eg. MLJ.jl) you can wrap using Tables.table(X) or Tables.table(adjoint(X)) as appropriate. The Tables.table method is provided by Tables.jl

I’m not aware of any Julia ML packages that specifically tailor algorithms to MltArrays, but they may exist.

3

Metal does not have FFT, so no cnn?