Compare trained and evaluated models: Notebook

As a Data Scientist, this is a common workflow: Train a model locally (in my Notebook), log the parameters, log the training time series metrics to Vertex AI TensorBoard, and log the evaluation metrics.

You can view the experiment runs associated with an experiment on the experiments page in the Cloud de Confiance console.

Notebook: Compare locally trained models

In the "Gemini Enterprise Agent Platform: Track parameters and metrics for locally trained models" notebook, you'll learn how to use Vertex AI Experiments to:

• Log the model parameters.
• Log the loss and metrics on every epoch to TensorBoard.
• Log the evaluation metrics.
• Compare two experiment runs.

Relevant content

• Log data to an experiment run
• Assign backing Vertex AI TensorBoard resource for Time Series Metric
• Log summary metrics
• Log time series metrics
• Log parameters