CommEfficient
PyTorch for benchmarking communication-efficient distributed SGD optimization algorithms
Install / Use
/learn @kiddyboots216/CommEfficientREADME
CommEfficient
This repo uses PyTorch's native distributed module in an implementation of federated learning.
It comes with a few experimental setups; various Residual Networks on CIFAR10, CIFAR100, FEMNIST, ImageNet (cv_train.py) and GPT2 on PersonaChat (gpt2_train.py).
There are a variety of command-line args which are best examined by looking at utils.py
The server is contained in fed_aggregator.py and the worker is contained in fed_worker.py
Other relevant branches: attacks contains an implementation of a malicious model poisoning adversary.
To use sketching, you need to install https://github.com/nikitaivkin/csh
Related Skills
node-connect
343.3kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
92.1kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
343.3kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
343.3kQQBot 富媒体收发能力。使用 <qqmedia> 标签,系统根据文件扩展名自动识别类型(图片/语音/视频/文件)。
