AWS re:Invent 2019: [NEW LAUNCH!] Amazon SageMaker Debugger: Insights into ML model (AIM216-R1)

Published on Dec 10, 2019

During ML model training, it’s challenging to ensure that models are progressively learning the correct values for different parameters and to analyze and debug model characteristics without building additional tools, making the process time-consuming and cumbersome. Come learn about Amazon SageMaker Debugger, a new capability that provides complete insights into the training process by automating data capture and analysis from training runs without code changes. Learn how you can analyze data using the Amazon SageMaker Studio visual interface and be alerted when anomalies and errors are detected, reducing the time needed to debug models from days to minutes. Amazon SageMaker Debugger helps you solve problems quickly, reduce troubleshooting time during training, and build high-quality models.


2020