TensorBoard provides tooling for tracking and visualizing metrics as well as visualizing models. All repositories that contain TensorBoard traces have an automatic tab with a hosted TensorBoard instance for anyone to check it out without any additional effort!
Over 6,000 repositories have TensorBoard traces on the Hub. You can find them by filtering at the left of the models page. As an example, if you go to the pyannote/embedding repository, there is a Metrics tab. If you select it, you’ll view a TensorBoard instance.
THe Hub automatically detects TensorBoard traces (such as
tfevents). Once you push your TensorBoard files to the Hub, they will automatically start an instance.
- TensorBoard documentation.