After more than three and a half years in the Apache incubator, Singa fulfills all the conditions of an Apache project, as the Apache Software Foundation announces. Apache Singa is a distributed, scalable machine learning library. Singa was developed by the National University of Singapore in 2014 and handed over to the Apache Software Foundation (ASF) in March 2015. Three years ago, the first version of Singa, version 0.1.0 was released. Further releases followed up to version 2.0.0 half a year ago. The first release as an official Apache project is still pending.
Singa uses a variety of well-known distributed training strategies, including synchronous and asynchronous training. According to the project, Singa’s users include, among others, Carnegie Technologies, CBRE, Citigroup, the National University of Singapore, NetEase, Noblis, Shentilium Technologies YZBigData, and various clinics.
Submitted by: Arnfried Walbrecht