MXNet Model Server Documentation

Basic Features

• Serving Models - How to use mxnet-model-server
  • REST API - Specification on the API endpoint for MMS
  • Model Zoo - A collection of MMS .model files that you can use with MMS
• Exporting Models - How to use mxnet-model-export
• Docker - How to use MMS with Docker and cloud services

Advanced Features

• Custom Model Service - Custom inference services and dev guide for implementation
• Client Code Generation - Use Swagger to create a client API for over 40 languages; a JavaScript example is provided
• Unit Tests - Housekeeping unit tests for MMS
• Load Test - Use JMeter to run MMS through the paces and collect load data

Examples

• LSTM - An example MMS project for an RNN / LSTM that will take json inputs for inference against a model trained with a specific vocabulary
• Object Detection - An example MMS project for using a pre-trained Single Shot Multi Object Detection (SSD) model taking image inputs and inferring the types and locations of several classes of objects