A Flexible and Powerful Parameter Server for large-scale machine learning

Lightweight and Scalable framework that combines mainstream algorithms of Click-Through-Rate prediction based computational DAG, philosophy of Parameter Server and Ring-AllReduce collective communication.

extremely distributed machine learning

自己实现的深度学习训练框架，纯java实现，没有过多的第三方依赖，可分布式训练

Distributed Fieldaware Factorization Machines based on Parameter Server

A parameter server implement with MPI.

DDLS is a parameter server based Distributed Deep Learning Studio for training deep learning models on Big Data with a numbers of machines and deploying high-performance online model service

分布式流数据大规模主题模型的实现

Serving layer for large machine learning models on Apache Flink

Machine Learning models for large datasets

a simple machine learning library

A lightweight parameter server interface

A simple and basic implement of parameter server for caffe.

A lightweight community-aware heterogeneous parameter server paradigm.

A parameter server compatible with PyTorch optimizers.

parameter server

Improving Performance for Distributed SGD using Ray