DeepSpeed
DeepSpeed is an open source deep learning optimization library for PyTorch. The library is designed to reduce computing power and memory use and to train large distributed models with better parallelism on existing computer hardware. DeepSpeed is optimized for low latency, high throughput training. It includes the Zero Redundancy Optimizer for training models with 100 billion parameters or more. Features include mixed precision training, single-GPU, multi-GPU, and multi-node training as well as custom model parallelism. The DeepSpeed source code is licensed under MIT License and available on GitHub.
