First Steps AWS Serverless for Deep Learning Applications
How to use AWS serverless infrastructure to organize deep learning workflows for your applications
Deep learning is essential for many companies. But a common issue—after the initial success of building a model—is how to organize the workflows for deep learning training and inference. Serverless architecture changes the rules of the game, providing a cheap, simple, scalable, and reliable architecture for deep learning applications. Instead of thinking about cluster management, scalability, and queue processing, you can now focus entirely on training the model. The challenge with this approach is that you have to keep in mind certain limitations; how you organize training and inference of your model is critical.
Join expert Rustem Feyzkhanov to learn how to use services like AWS Batch, AWS Fargate, AWS Sagemaker, AWS Lambda, and AWS Step Functions to organize deep learning workflows. You’ll explore common challenges you may encounter in your deep learning applications—and discover how to solve them.
What you'll learn-and how you can apply it
By the end of this live online course, you’ll understand:
- Common challenges you’ll face when building deep learning applications
- How to tackle them using serverless infrastructure
- How to apply serverless infrastructure to your use case
And you’ll be able to:
- Deploy spot GPU instances for deep learning training using AWS Batch
- Deploy scalable deep learning inference using AWS Lambda
- Organize deep learning workflows for AI applications using AWS Step Functions
This training course is for you because...
- You’re a data scientist or data analyst.
- You work with TensorFlow or PyTorch.
- You want to become a machine learning engineer.
- Familiarity with AWS services
- A basic understanding of TensorFlow
- A free AWS account for testing (To do the exercises, you’ll need to input AWS account credentials. For your protection, please do not use credentials from a “real” AWS account that is used for any other purpose.)
- Watch Practical Deep Learning on the Cloud (video, 2h 27m)
- Watch Serverless Deep Learning with TensorFlow and AWS Lambda (video, 1h 29m)
About your instructor
Rustem Feyzkhanov is a machine learning engineer at Instrumental, where he creates analytical models for the manufacturing industry. Rustem is passionate about serverless infrastructure (and AI deployments on it) and is the author of the course and book "Serverless Deep Learning with TensorFlow and AWS Lambda" and "Practical Deep Learning on the Cloud".
The timeframes are only estimates and may vary according to how the class is progressing
Using serverless infrastructure for deep learning applications (60 minutes)
- Presentation: Introduction to ML tasks and challenges; introduction to AWS serverless infrastructure—AWS Batch, AWS Fargate, AWS Lambda, AWS SageMaker, and AWS Step Functions; using AWS Batch and AWS Sage Maker for training; using AWS Lambda and AWS Step Functions for deep learning inference
- Hands-on exercises: Deploy AWS Step Functions with AWS Lambda using a serverless framework; deploy AWS Step Functions with AWS Batch using a serverless framework; deploy AWS Step Functions with AWS Lambda for deep learning inference using a serverless framework