TensorBoard

+ Gradient

How to use Gradient and

TensorBoard

together

Use the TensorBoard integration to visualize training performance on Gradient.

TensorBoard is a popular open source visualization software that comes with any standard TensorFlow installation but is a first class citizen in other frameworks such as PyTorch. TensorBoard is natively supported on the Gradient platform. Within a project you can easily provision an TensorBoard instance with the web interface or CLI. From there you can attach Experiments (training runs) to the TensorBoard to visualize and compare your training iterations.

Creating and managing TensorBoards in Gradient

Within a Gradient project, you can create TensorBoards, add experiments to them, and remove experiments from them. You can do all of this via the Web UI or the CLI.

Gradient offers push-button support for TensorBoard

Manage multiple TensorBoards with a simple interface

You'll see all of your team's TensorBoards listed under any Experiment Details view, though you can filter TensorBoards down only to show the ones that contain Experiments from within the current Project. Learn more about the Gradient TensorBoard integration in the docs.