Learn how to monitor AI models in real-time using the Comet ML platform in this 27-minute tutorial. Explore key features of model monitoring, including tracking metrics associated with risk and drift. Follow a step-by-step guide to log models, register experiments, track model versions, and deploy registered models. Dive into a practical demonstration using a heart disease detection experiment in Keras, and discover how to set up and configure the Comet ML Python package. Gain insights into visualizing monitoring data, charts, and histograms. Master experiment logging configuration and learn how to save experiments to GitHub. Perfect for AI engineers looking to enhance their model monitoring skills and leverage the power of Comet ML for improved model performance tracking in production environments.
An AI Engineer Guide to Model Monitoring with Comet ML Platform
Overview
Syllabus
- Video Start
- Video Content Intro
- Demo Notebook Introduction
- Sample notebook Tutorial
- Heart Disease Detection in Keras Experiment
- Comet ML Platform access and API Key
- Comet ML Dashboard
- Comet ML Python Package Installation
- Comet ML Python Package Configuration
- Comet ML Python Package Initialization
- Experiment Monitoring Data
- Monitoring Data and Charts Visualization
- Histogram View
- Experiment logging configuration
- Saving Experiments to GitHub
- Recap
- Credits
Taught by
Prodramp
