Quick docker container that serves up tensorboard over tensorflow/tensorflow:latest
The build file should build a docker image of the container (and pull down the latest tensorflow docker base). You will need a network connection to use this. Also it might require running sudo, be sure to read the simple bash script before running.
The resulting image will be called rthosfelt_tensorboard and the run script relies on that, change both image names if you know what you're doing.
The run script takes in a directory to mount to the container as the /logs/ directory and will be updated as logs are added (even under subfolders). Again, you might be required to run this with sudo be careful!
This software is distributed under the terms of both the MIT license and the Apache License (Version 2.0), no Guarantee or promises provided. READ THE SCRIPT before running!
See LICENSE-APACHE and LICENSE-MIT for details.