Value: $126.99
(as of Sep 14, 2025 15:40:33 UTC – Particulars)
Take your NLP information to the following stage and grow to be an AI language understanding professional by mastering the quantum leap of Transformer neural community fashions
Key Options
Construct and implement state-of-the-art language fashions, akin to the unique Transformer, BERT, T5, and GPT-2, utilizing ideas that outperform classical deep studying modelsGo by means of hands-on purposes in Python utilizing Google Colaboratory Notebooks with nothing to put in on an area machine Take a look at transformer fashions on superior use circumstances
Ebook Description
The transformer structure has proved to be revolutionary in outperforming the classical RNN and CNN fashions in use as we speak. With an apply-as-you-learn strategy, Transformers for Pure Language Processing investigates in huge element the deep studying for machine translations, speech-to-text, text-to-speech, language modeling, query answering, and lots of extra NLP domains with transformers.
The e book takes you thru NLP with Python and examines varied eminent fashions and datasets throughout the transformer structure created by pioneers akin to Google, Fb, Microsoft, OpenAI, and Hugging Face.
The e book trains you in three phases. The primary stage introduces you to transformer architectures, beginning with the unique transformer, earlier than shifting on to RoBERTa, BERT, and DistilBERT fashions. You’ll uncover coaching strategies for smaller transformers that may outperform GPT-3 in some circumstances. Within the second stage, you’ll apply transformers for Pure Language Understanding (NLU) and Pure Language Era (NLG). Lastly, the third stage will show you how to grasp superior language understanding methods akin to optimizing social community datasets and faux information identification.
By the tip of this NLP e book, you’ll perceive transformers from a cognitive science perspective and be proficient in making use of pretrained transformer fashions by tech giants to varied datasets.
What You Will Study
Use the newest pretrained transformer modelsGrasp the workings of the unique Transformer, GPT-2, BERT, T5, and different transformer modelsCreate language understanding Python applications utilizing ideas that outperform classical deep studying modelsUse a wide range of NLP platforms, together with Hugging Face, Trax, and AllenNLPApply Python, TensorFlow, and Keras applications to sentiment evaluation, textual content summarization, speech recognition, machine translations, and moreMeasure the productiveness of key transformers to outline their scope, potential, and limits in manufacturing
Who this e book is for
For the reason that e book doesn’t educate primary programming, you should be accustomed to neural networks, Python, PyTorch, and TensorFlow with a view to be taught their implementation with Transformers.
Readers who can profit probably the most from this e book embrace skilled deep studying & NLP practitioners and information analysts & information scientists who wish to course of the rising quantities of language-driven information.
Writer : Packt Publishing
Publication date : Jan. 29 2021
Language : English
Print size : 384 pages
ISBN-10 : 1800565798
ISBN-13 : 978-1800565791
Merchandise weight : 210 g
Dimensions : 19.05 x 2.21 x 23.5 cm
Greatest Sellers Rank: #1,081,647 in Books (See High 100 in Books) #386 in Pure Language Processing Software program #397 in AI Human Imaginative and prescient & Language Methods #398 in A.I. Neural Networks
Buyer Evaluations: 4.3 4.3 out of 5 stars 88 rankings var dpAcrHasRegisteredArcLinkClickAction; P.when(‘A’, ‘prepared’).execute(operate(A) { if (dpAcrHasRegisteredArcLinkClickAction !== true) { dpAcrHasRegisteredArcLinkClickAction = true; A.declarative( ‘acrLink-click-metrics’, ‘click on’, { “allowLinkDefault”: true }, operate (occasion) { if (window.ue) 0) + 1); } ); } }); P.when(‘A’, ‘cf’).execute(operate(A) { A.declarative(‘acrStarsLink-click-metrics’, ‘click on’, { “allowLinkDefault” : true }, operate(occasion){ if(window.ue) 0) + 1); }); });
