Deep Learning (MIOpen)
AMDGPU.jl exposes AMD's MIOpen deep-learning primitives through the AMDGPU.MIOpen submodule: convolutions, pooling, softmax, batch normalization, and activation functions, all operating on ROCArray.
In practice you rarely call these directly. They back the GPU implementations used by NNlib.jl and Flux.jl, so building and training a model on an AMD GPU with Flux uses MIOpen automatically once your data and parameters are ROCArrays.
Check that MIOpen is available with:
julia
using AMDGPU
AMDGPU.functional(:MIOpen)If it reports false, see Installation: the downloaded artifacts include MIOpen, a local ROCm needs its MIOpen package installed.