1 repo
Awesome GitHub RepositoriesGradient Management Techniques
Methods for selecting, filtering, or prioritizing gradient updates during the training process.
Distinguishing note: Focuses on selective update strategies rather than full-state compression.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Gradient Management Techniques. Refine with filters or upvote what's useful.
Awesome Gradient Management Techniques GitHub Repositories
Find the best repos with AI.We'll search the best matching repositories with AI.
- deepspeedai/DeepSpeed
deepspeedai/DeepSpeed
41,638View on GitHubDeepSpeed is a high-performance library designed to scale deep learning model training and inference across massive clusters of GPUs and compute nodes. It provides a comprehensive suite of tools for distributed training, enabling the execution of models that exceed the memory capacity of single devices through advanced parameter partitioning, pipeline-based model parallelism, and memory-efficient state offloading. The framework distinguishes itself through specialized communication-efficient optimizers and hardware-aware acceleration techniques. By utilizing gradient compression, quantization
The framework reduces communication overhead by updating only critical gradients during training steps while offloading remaining computations to CPU memory.
Pythonbillion-parameterscompressiondata-parallelism
