smatovic wrote: ↑Sat Jan 07, 2023 9:31 amThe effort will be called the Unified AI Stack, and the first version will bring together AMD's ROCm software for GPU programming, its CPU software and Xilinx's Vitis AI software.
Given that AMD still struggles with basic driver problems, I wouldn't bet a dime on their software stack. Also, AMD has missed the future for so long that their only hope to get anywhere is at least source level / API compatibibility with CUDA/ML.
True. If we look at the AMD GPGPU software stack history since 2008 and the one from Nvidia....and now AMD has HIP for CUDA compatibility, maybe Vulkan with SPIR-V as intermediate layer will change something in future, you can cross-compile OpenCL to run on Vulkan devices via SPIR-V, and almost all modern GPUs speak Vulkan, even the Raspberry Pi 4.
Henk wrote: ↑Sat Jan 07, 2023 10:53 am
I don't like neural networks at all. If you have no clue you use neural networks. But that is similar to resignation.
Neural networks will never be 100% accurate. Being statistical machines and black boxes.
There you have your task, un-black box the black box
