You're reading from Pretrain Vision and Large Language Models in Python End-to-end techniques for building and deploying foundation models on AWS

Product type Paperback

Published in May 2023

Publisher Packt

ISBN-13 9781804618257

Length 258 pages

Edition 1st Edition

Languages

Python

Tools

AWS

Concepts

GPT/LLMs

Author (1):

Emily Webber

View More author details

Table of Contents (23) Chapters

Preface

1. Part 1: Before Pretraining

2. Chapter 1: An Introduction to Pretraining Foundation Models FREE CHAPTER

3. Chapter 2: Dataset Preparation: Part One

4. Chapter 3: Model Preparation

5. Part 2: Configure Your Environment

6. Chapter 4: Containers and Accelerators on the Cloud

7. Chapter 5: Distribution Fundamentals

8. Chapter 6: Dataset Preparation: Part Two, the Data Loader

9. Part 3: Train Your Model

10. Chapter 7: Finding the Right Hyperparameters

11. Chapter 8: Large-Scale Training on SageMaker

12. Chapter 9: Advanced Training Concepts

13. Part 4: Evaluate Your Model

14. Chapter 10: Fine-Tuning and Evaluating

15. Chapter 11: Detecting, Mitigating, and Monitoring Bias

16. Chapter 12: How to Deploy Your Model

17. Part 5: Deploy Your Model

18. Chapter 13: Prompt Engineering

19. Chapter 14: MLOps for Vision and Language

20. Chapter 15: Future Trends in Pretraining Foundation Models

21. Index

Why subscribe?

22. Other Books You May Enjoy

Troubleshooting accelerator performance

Before we can analyze our GPU performance, we need to understand generally how to debug and analyze performance on our training platform. SageMaker has some really nice solutions for this. First, all of your logs are sent to Amazon CloudWatch, another AWS service that can help you monitor your job performance. Each node in your cluster will have a full dedicated log stream, and you can read that log stream to view your overall training environment, how SageMaker runs your job, what status your job is in, and all of the logs your script emits. Everything you write to standard out, or print statements, is automatically captured and stored in CloudWatch. The first step to debugging your code is to take a look at the logs and figure out what really went wrong.

Once you know what’s wrong in your script, you’ll probably want to quickly fix it and get it back online, right? That’s why we introduced managed warm pools on SageMaker...