Originally published at: Automatic Mixed Precision Now Natively Available in MXNet | NVIDIA Technical Blog
NVIDIA and Apache MXNet partner to simplify mixed precision training in MXNet. Today, Apache MXNet announced native support for NVIDIA’s Automatic Mixed Precision (AMP) training feature on Volta and Turing GPUs. Developers can now easily access deep learning training speedups available from NVIDIA Tensor Cores using reduced precision. “Designed specifically for deep learning, the first-generation…