Produto disponível em até 15min no aplicativo Kobo, após a confirmação  do pagamento!
Você pode ler este livro digital em vários dispositivos:
Understanding and coding advanced deep learning algorithms with the most intuitive deep learning library in existence
About This Book
A complete and up-to-date introduction to GANs
A complete overview of Keras
A dive into advanced deep learning sticking to the essential mathematics
Who This Book Is For
Familiarity with Python and basic machine learning is necessary for this book and it would be preferable if the reader had understood several basic deep learning algorithms, like CNNs and RNNs.
What You Will Learn
You will learn Keras thoroughly
To code image synthesis examples with GANs
To distinguish various types of adversarial networks and implement them
To write a reinforcement learning application with OpenAI gym
To step away from classic deep learning and machine learning and write production-ready applications based on recent research
Keras enables a new generation of deep learning developers to access the full power of TensorFlow on the one hand, while concentrating on building applications on the other. Even more surprising is the ability to write applications drawing from the power of new algorithms, without actually having to implement all the algorithms, since they are already available.
After introducing Keras and familiarizing the reader with Keras via classical deep learning algorithms, Dr. Atienza walks the developer through autoencoders first. He takes the approach of building on relatively well-known approaches and algorithms, before introducing more recent developments to working developers. He then asks the reader to write and understand an NLP application to prove the practical value of autoencoders written with Keras.
The core of the book lies in the coverage of several classes of adversarial networks (GANs). Dr. Atienza focuses on the most recent successes of GANs and teaches developers to implement newer results for themselves, warning them of pitfalls and showing them the advantages of each. He focuses in particular on Image generation and synthesis.
Finally, the book finishes with an introduction to reinforcement learning, using OpenAI Gym as a framework to simplify experimenting with various policies and algorithms. Again, Keras is the unifying layer through which OpenAI Gym is accessed.
Overall, Advanced Keras shows the full capabilities of Keras to a developer, while trying to avoid looking at underlying infrastructure provided by TensorFlow, Theano or Microsoft Cognitive Services. Dr. Atienza is showing how to get new algorithms to work within Keras, without getting the reader tangled in too many implementation details.