Long Short-Term Memory Networks With Python: Develop Sequence Prediction Models With Deep Learning

by Jason Brownlee

The Long Short-Term Memory network, or LSTM for short, is a type of recurrent neural network that achieves state-of-the-art results on challenging prediction problems. In this laser-focused Ebook, finally cut through the math, research papers and patchwork descriptions about LSTMs. Using clear explanations, standard Python libraries and step-by-step tutorial lessons you will discover what LSTMs are, and how to develop a suite of LSTM models to get the most out of the method on your sequence prediction problems.

Genres: Artificial Intelligence, Computer Science

246 pages, ebook

Published January 1, 2017

About the author

Jason Brownlee, Ph.D. trained and worked as a research scientist and software engineer for many years (e.g. enterprise, R&D, and scientific computing), and is known online for his work on Computational Intelligence (e.g. Clever Algorithms), Machine Learning and Deep Learning (e.g. Machine Learning Mastery, sold in 2021) and Python Concurrency (e.g. Super Fast Python).

Jason writes fiction under the pseudonym J.D. Brownlee: https://www.goodreads.com/jdbrownlee

Reviews

[Damian]

'Long Short-Term Memory Networks With Python' is excelent book if
you want to learn programming LSTM networks in tensorflow and keras.

Author starts with basic knowledge and supoprt it with excelent
examples - both theoretical and in python code. It's very easy to understand and in
most cases you can copy/paste them and see how they are working on your own
machine.

I was reading a lot of articles online about LSTM's, the knowledge
is scattered and it's hard to focus on particular aspect of this topic. This
book helped me sort everything out and finally I was able to focus - I have
everything in one place, well documented and with working code.

This book is divided in nice logical way - author exaplain what
LSTM networks are and how you can use them, then you have nice chapter about LSTM basics,
after that you have exampels of actual models and in the end we have advanced topics -
knowledge in this chapter can also be used in other Machine Learning techniques - not only
LSTM's.

For me the most valuable chapters of this book are:
- Encoder-Decoder LSTM's
- Bidirectional LSTM's
- Sequence Predictions with LSTM's

After reading this book I was able to grasp bacis understanding
about those topics.

I can highly recommend this book. If you don't have time to look
informations about LSTM's online and you value your time, this is the book for you. You can
also buy this books for code examples alone - they are worth the price.

[Bing Wang · Rating 4 out of 5]

Good instruction book if you:
1. know the theory of LSTMs, what is input gate, update gate, forget gate, hidden state, cell state etc.
2. know how deep learning model is structured and optimized, layers to layers, SGD, Adam etc
3. know basic deep learning modeling tricks. Like batch learning, early stopping, dropout etc.
4. want to learn what LSTM could, could not do
5. how to construct LSTMs with mature tools (not build from earth, you do not have to code your LSTM cells)
6. want to learn how problem is formulated into deep learning framework with LSTMs
7. know basic operations of numpy, pandas and scipy

[Mobina golsefatan · Rating 4 out of 5]

اگر آشنایی مختصری با پایتون و شبکه عصبی داشته باشید این کتاب یهتون کمک می‌کنه که بتونید کدهای LSTM رو متوجه بشید
. اما هنوز نمی‌تونم درباره‌ی کاربردی بودنش نظر بدم. این بماند برای بعد از اتمام پروژه.
در کل توضیحات خوب و قابل فهمه.