Introduction

This proposal is intended for supporting MXNet data visualization in the TensorFlow's TensorBoard. We plan to develop a logging tool bundled in MXNet python package for users to log data in the format that the TensorBoard can later render in browsers. The typical flow of using the logging tool is described in the following figure. Users would need to install MXNet and TensorFlow's TensorBoard to visualize the data. The project will be divided into two phases: 1. Synchronized logger; 2. Asynchronized logger. Now let's discuss the design details in the following sections.