Uber today brought Horovod, a framework for distributed training across multiple machines, to open source initiative LF Deep Learning Foundation. Uber has used Horovod to support self-driving vehicles, fraud detection, and trip forecasting. Contributors to the project include Amazon, IBM, Intel, and Nvidia.
In addition to Uber, Alibaba, Amazon, and Nvidia also use Horovod.
The Horovod project can be used with popular frameworks like TensorFlow, Keras, and PyTorch.
Uber joined Linux Foundation last month and joins other tech companies like AT&T and Nokia that have come forward to support LF Deep Learning Foundation open source projects.
The announcement was made at KubeCon + CloudNativeCon North America being held this week in Seattle.
The LF Deep Learning Foundation was created in March to support open source initiatives for deep learning and machine learning and is part of the Linux Foundation.
The launch of Horovod comes a month after the launch of Acumos, for training and deploying AI models, and the Acumos Marketplace, an open exchange for AI models.
Other projects undertaken since the launch of the foundation include the machine learning platform Angel and Elastic Deep Learning, a project to help cloud service providers make cloud cluster services with frameworks like TensorFlow.
Those projects were added in August by Baidu and Tencent respectively, which are founding members of the LF Deep Learning Foundation.