Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Save more on your purchases! discount-offer-chevron-icon
Savings automatically calculated. No voucher code required.
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Newsletter Hub
Free Learning
Arrow right icon
timer SALE ENDS IN
0 Days
:
00 Hours
:
00 Minutes
:
00 Seconds
Arrow up icon
GO TO TOP
Natural Language Processing with Flair

You're reading from   Natural Language Processing with Flair A practical guide to understanding and solving NLP problems with Flair

Arrow left icon
Product type Paperback
Published in Apr 2022
Publisher Packt
ISBN-13 9781801072311
Length 200 pages
Edition 1st Edition
Languages
Tools
Arrow right icon
Author (1):
Arrow left icon
Tadej Magajna Tadej Magajna
Author Profile Icon Tadej Magajna
Tadej Magajna
Arrow right icon
View More author details
Toc

Table of Contents (15) Chapters Close

Preface 1. Part 1: Understanding and Solving NLP with Flair
2. Chapter 1: Introduction to Flair FREE CHAPTER 3. Chapter 2: Flair Base Types 4. Chapter 3: Embeddings in Flair 5. Chapter 4: Sequence Tagging 6. Part 2: Deep Dive into Flair – Training Custom Models
7. Chapter 5: Training Sequence Labeling Models 8. Chapter 6: Hyperparameter Optimization in Flair 9. Chapter 7: Train Your Own Embeddings 10. Chapter 8: Text Classification in Flair 11. Part 3: Real-World Applications with Flair
12. Chapter 9: Deploying and Using Models in Production 13. Chapter 10: Hands-On Exercise – Building a Trading Bot with Flair 14. Other Books You May Enjoy

Knowing when to stop and try again

The question "Is my model good enough?" is a whole subfield of machine learning on its own and can't fully be explained in a single chapter. Nevertheless, there's still some practical advice we all can follow to help us determine whether it's worth restarting training with different parameters, different model types, or even new data.

The two basic techniques for measuring success during and after training are as follows:

  1. Monitoring loss
  2. Assessing and comparing performance metrics

Monitoring loss

The first and most important thing you should look out for during training is whether the model is learning in an expected way. This is done by monitoring training output metrics such as loss. A typical training session follows the following pattern. In the initial stage of training, our weights are randomized, and loss will be huge (remember how bad our model with random weights was?). Then, even after one...

lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime
Banner background image