liveBook · Manning

Comet

As explained in Data Storytelling with Altair and AI

Comet is a machine-learning experiment management platform for tracking experiments, models, and evaluation results.

Comet is an experimentation platform for tracking and testing machine-learning models. It also combines experiment results, data stories, and reports in one workflow. section 11.5

Use Comet when an Altair data story belongs to a broader data-science project involving model experimentation and evaluation. Python support lets you continue working in Python while publishing reports. section 11.5 section 11.8

A typical Comet publishing workflow is:

  • Create a project configured for experiment tracking. section 11.5
  • Create a report and add a new panel. section 11.5
  • Choose Python or JavaScript, add Vega, Vega-Lite, and Vega-Embed resources, and render the chart in the panel. section 11.5

Embedding an interactive chart

A slider.html visualization can become a Comet report panel. The panel loads Vega, Vega-Lite, and Vega-Embed, targets the vis element, and renders the chart specification interactively. section 11.5

How Data Storytelling with Altair and AI covers it

How deep it goes
Demonstrates project and report setup plus interactive HTML/JavaScript chart embedding, while deferring broader Comet usage.
Examples
Uses a slider.html interactive chart, Comet report panels, and Vega-Embed rendering code.
What it assumes
Assumes an Altair-produced data story and a data-science project containing model experimentation and evaluation.
What you can do after
Apply it via Comet to publish interactive charts alongside tracked experiments and evaluation results.
How it connects
Connects data storytelling with Altair, Vega, Vega-Lite, JavaScript, Python, and experiment tracking.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Data Storytelling with Altair and AI. The section links point to the parts of the book it draws on.