Learn to build, train, and iterate machine learning models faster with new AWS course

Did you know ML played a key role in reducing the number of potential COVID vaccine candidates from tens of thousands to 26 (PMC)? Improving productivity is the key to delivering value quickly, and in the case of COVID vaccine development, it saved lives.

For experienced data scientists working with disparate data science tools, Amazon SageMaker Studio provides an integrated set of ML tools in a single interface. Our new three-day, advanced-level, virtual classroom course, Amazon SageMaker Studio for Data Scientists, will help you develop the skills to increase productivity at every stage of the ML lifecycle using SageMaker Studio.

What is different about machine learning today?

The unstructured data analytics and data management market in the public cloud is expected to grow at a CAGR of 41.9% between 2021 and 2025 (IDC). Since data is at the core of ML activities, we can also expect ML model building activity to move to the cloud to reduce data movement and to speed up model development. If you already have data and analytics workloads on AWS, or are considering AWS Cloud services, you can improve model building performance by consolidating ML workloads on AWS Cloud.

About Amazon SageMaker Studio

Amazon SageMaker Studio is a purpose-built Integrated Development Environment (IDE) for ML. SageMaker Studio performs all ML development steps, from preparing raw data to deploying and monitoring ML models, with access to the most comprehensive set of tools in a single web-based visual interface. You can build models by accessing data from multiple sources including Amazon Simple Storage Service (Amazon S3), Amazon Redshift, and Snowflake.

The respondents in the 2020 Anaconda survey said 66% of their time was spent loading, cleansing, and visualizing data. As data continues to grow exponentially, tools to transform, analyze, and visualize data faster become even more important. For example, SageMaker Data Wrangler, integrated with SageMaker Studio, reduces the data transformation time with the help of 300+ built-in functions and generates ML-powered insights. SageMaker Studio also supports data processing at scale through notebook integration with Apache Spark running on Amazon EMR clusters. Data scientists also have the option to use the serverless Apache Spark runtime environment managed by AWS Glue Interactive Sessions to interactively prepare data at scale right in their Studio notebooks.

SageMaker Studio improves data scientists’ productivity by automatically tracking and charting details related to experiments and trials, in addition to providing model debugging and profiling help with SageMaker Debugger. Using SageMaker Clarify data scientists can identify biases in data and models, and get insights into the reasons for certain predictions (explainability).

Developing the skills needed to take advantage of these capabilities is critical to organizations migrating from on-premises machine learning to AWS Cloud, and for customers building cloud native solutions with SageMaker.

About the three-day classroom course

Amazon SageMaker Studio for Data Scientists is an advanced-level, three-day course teaching you how to build, train, and iterate ML models faster using SageMaker Studio in a hands-on-lab environment with the help of expert AWS instructors.

You’ll learn five major time-saving skills:

1. How to engineer features using built-in transformations in SageMaker Data Wrangler and share those features using SageMaker Feature Store;
2. How to build models faster using built-in algorithms, SageMaker Autopilot, SageMaker Debugger, and automatic model tuning;
3. How to compare the performance of various trials associated with model training using SageMaker Experiments, and track them in SageMaker Model Registry;
4. How to identify biases in data and model using SageMaker Clarify; and,
5. How to automate the model building workflow using SageMaker Pipelines.

The course follows the ML lifecycle starting with feature engineering, progressing to model building, training, and tuning, followed by deployment, inference, and monitoring. You’ll learn eight major features of Amazon SageMaker Studio with the help of 10 labs.

In addition, AWS instructors will use interactive sessions to walk you through SageMaker Studio User Interface (UI), SageMaker Autopilot, and SageMaker Model Monitor. Finally, you’ll be given seven challenges in one day to test your understanding of SageMaker Python SDK and SageMaker Studio. You can select the level of assistance for each challenge – no assistance, hints only, or solution walk-through.

To get the most out of this course we recommend learners have one+ year of ML experience and foundational knowledge of AWS. You can satisfy the foundational knowledge requirement by completing the AWS Technical Essentials course.

Whether you attend the class virtually or in-person, you’ll have the opportunity to ask questions, work through solutions with your peers, and get real-time feedback from accredited AWS instructors with deep technical knowledge.

Is the AWS Certified Machine Learning – Specialty your goal?

If you want to earn an industry-recognized credential from AWS that validates your expertise in AWS Machine Learning, you may want to consider the AWS Certified Machine Learning – Specialty certification. While the Amazon SageMaker Studio for Data Scientists course explores SageMaker Studio-centered data processing, model building, training, tuning, and pipeline topics, we offer additional information to help you prepare for the AWS Certified Machine Learning – Specialty exam.

What resources are available if I want to learn more?

If you’re interested in learning more about our AWS Training and Certification offerings for ML, download our AWS Machine Learning Ramp-Up Guide. We offer many free, on-demand digital resources as well as several virtual instructor-led courses for machine learning.