Model Server for Apache MXNet Introduces ONNX Support and Amazon CloudWatch Integration
Version 0.2 of Model Server for Apache MXNet (MMS), an open source library for Apache MXNet, is now available for packaging and serving deep learning models for inference at scale. You can use MMS to serve ONNX models created with any ONNX-supporting deep learning framework, such as PyTorch, Caffe2, Microsoft Cognitive Toolkit, or Chainer. ONNX, the Open Neural Network Exchange format, is a community initiative driven by AWS, Facebook, and Microsoft, with growing support across additional deep learning frameworks and platforms. To get started serving ONNX models, see MMS ONNX Serving documentation.
Version 0.2 of Model Server for Apache MXNet (MMS), an open source library for Apache MXNet, is now available for packaging and serving deep learning models for inference at scale. You can use MMS to serve ONNX models created with any ONNX-supporting deep learning framework, such as PyTorch, Caffe2, Microsoft Cognitive Toolkit, or Chainer. ONNX, the Open Neural Network Exchange format, is a community initiative driven by AWS, Facebook, and Microsoft, with growing support across additional deep learning frameworks and platforms. To get started serving ONNX models, see MMS ONNX Serving documentation.
This new MMS release also includes integration with the Amazon CloudWatch API, which makes publishing operational metrics from MMS into CloudWatch easy and convenient. With this integration, engineers benefit from CloudWatch’s web-based dashboard, metrics rendering in real time, and the ability to configure triggers and alerts. To get started with the CloudWatch integration for MMS, see the MMS CloudWatch Metrics documentation.
Modified 9/27/2021 – In an effort to ensure a great experience, expired links in this post have been updated or removed from the original post.