SageMaker Training and Deployment with Custom Images

How and Why to Use Custom SageMaker Images If you have used SageMaker for data science modeling work, you have likely used the AWS-provided images to train your models and possibly deploy them to an endpoint. These provide images for scikit-learn, xgboost and deep-learning among other common frameworks. This article will address an issue we […]

The Problem of Underfitting in Machine Learning

ChatGPT generated image What Underfitting is and How to Test for It and Minimize its Impact Machine learning stands as a pivotal element in contemporary data science, fundamentally altering the landscape of predictive analytics and decision-making across various domains. We have written many articles on machine learning concepts, but this one takes us back to […]

Data Quality Monitoring in AWS SageMaker

First things first, what is data quality monitoring? Data quality monitoring for machine learning can generally be thought of from two perspectives. One perspective is that of traditional data-engineering. This type of monitoring is concerned with the “physical” characteristics of the data and ensuring they are what you expect them to be. It involves criteria […]

Remote Development in Sagemaker Studio with VS Code

Disclaimer about Changes to Sagemaker Studio As of Nov. 30 2023, there have been major changes to Sagemaker Studio. Existing customers of Sagemaker Studio will get the default experience now called Sagemaker Studio Classic — this is the Studio experience this article was written for. New Sagemaker Studio customers (and existing customers that choose to […]