Tensorboard is a visualization tool that is packaged with Tensorflow. It is very useful for understanding and debugging Tensorflow projects. You can visualize your Tensorflow graph, plot quantitative metrics about the execution of your graph, and show additional data like images that pass through it.
Enabling Tensorboard for your project
To enable Tensorboard for your job run, you just need to add
--tensorboard flag to the floyd
run command. For example:
floyd run --env tensorflow-1.1 --tensorboard --mode jupyter
This should start a jupyter notebook using Tensorflow 1.1 environment and enable Tensorboard. You can open Tensorboard directly from the link on the job page in the dashboard.
You can view a full demo of this feature in this video:
As always, let us know if you have any feedback on this and other features on FloydHub.