Agent Platform TensorBoard is an enterprise-ready managed version of Open Source TensorBoard (TB), which is a Google Open Source project for machine learning experiment visualization.
With Agent Platform TensorBoard, you can track, visualize, and compare ML experiments and share them with your team.
Agent Platform TensorBoard provides various detailed visualizations, that includes:
- tracking and visualizing metrics such as loss and accuracy over time,
- visualizing model computational graphs (ops and layers),
- viewing histograms of weights, biases, or other tensors as they change over time,
- projecting embeddings to a lower dimensional space,
- and displaying image, text, and audio samples.
In addition to the powerful visualizations from TensorBoard, Agent Platform TensorBoard provides:
- a persistent, shareable link to your experiment's, Agent Platform TensorBoard experiment,
- tight integrations with Gemini Enterprise Agent Platform services for model training,
- enterprise-grade security, privacy, and compliance.
Integration with Colab Enterprise Experiments lets you:
- use a searchable and compare list of all experiments in a project,
- view time series metrics in the Google Cloud console,
- compare scalars across experiments and experiment runs,
- have direct access to Agent Platform TensorBoard.
Agent Platform TensorBoard and the Google Cloud console
The Google Cloud console is used to:
- create or delete Agent Platform TensorBoard instances,
- create Agent Platform Experiments,
- view Agent Platform TensorBoard instance storage size → associated costs,
- view associated Agent Platform Experiments, Custom Jobs, and Pipeline Runs,
- visualize some time series metrics.
What's next
- Setup Agent Platform TensorBoard
- Check out the TensorBoard API documentation.