You're reading from Mastering Transformers The Journey from BERT to Large Language Models and Stable Diffusion

Product type Paperback

Published in Jun 2024

Publisher Packt

ISBN-13 9781837633784

Length 462 pages

Edition 2nd Edition

Languages

Python

Tools

BERT

Concepts

GPT/LLMs

Authors (2):

Savaş Yıldırım

Meysam Asgari- Chenaghlu

View More author details

Table of Contents (25) Chapters

Preface

1. Part 1: Recent Developments in the Field, Installations, and Hello World Applications

2. Chapter 1: From Bag-of-Words to the Transformers FREE CHAPTER

3. Chapter 2: A Hands-On Introduction to the Subject

4. Part 2: Transformer Models: From Autoencoders to Autoregressive Models

5. Chapter 3: Autoencoding Language Models

6. Chapter 4: From Generative Models to Large Language Models

7. Chapter 5: Fine-Tuning Language Models for Text Classification

8. Chapter 6: Fine-Tuning Language Models for Token Classification

9. Chapter 7: Text Representation

10. Chapter 8: Boosting Model Performance

11. Chapter 9: Parameter Efficient Fine-Tuning

12. Part 3: Advanced Topics

13. Chapter 10: Large Language Models

14. Chapter 11: Explainable AI (XAI) in NLP

15. Chapter 12: Working with Efficient Transformers

16. Chapter 13: Cross-Lingual and Multilingual Language Modeling

17. Chapter 14: Serving Transformer Models

18. Chapter 15: Model Tracking and Monitoring

19. Part 4: Transformers beyond NLP

20. Chapter 16: Vision Transformers

21. Chapter 17: Multimodal Generative Transformers

22. Chapter 18: Revisiting Transformers Architecture for Time Series

23. Index

Why subscribe?

24. Other Books You May Enjoy

Faster inference using ONNX

Open Neural Network Exchange (ONNX) provides faster inference for models that are trained. The optimum library, on the other hand, provides easier ONNX exports for even pipelines of Hugging Face-based models. The usage and implementation are straightforward, as you will see:

The first thing to do is install the optimum and onnxruntime libraries:
```
$ pip install optimum[onnxruntime]
```

The next step is to load the pipeline using the optimum pipeline:

from optimum.pipelines import pipeline
pipe = pipeline("text-classification",
    "cardiffnlp/twitter-xlm-roberta-base-sentiment",
    accelerator="ort")

Two types of accelerators exist: ONNX runtime (ORT) and BETTERTRANSFORMER. ORT is used for the ONNX export of the model. For this specific example, we just picked a multilingual sentiment analysis model based on XLM-Roberta. Now that it has been converted, you can easily run the pipeline...