03967nam a22005055i 4500 ssj0002420662 WaSeSS 20220426142636.0 m d cr n 201129s2021 xxu| o |||| 0|eng d 9781484265123 10.1007/978-1-4842-6513-0 doi WaSeSS AIMIT LIBRARY Q334-342 UYQ bicssc COM004000 bisacsh UYQ thema 006.32 1 YALO EBOOK Yalcin, Orhan Gazi. 31958 Applied Neural Networks with TensorFlow 2 : API oriented deep learning with python / by Orhan Gazi Yalcin 1st ed. Berkeley, CA : Apress : Imprint: Apress, 2021. xix, 295 p. 23.4 cm. text file PDF rda Chapter 1: Introduction -- Chapter 2: Introduction to Machine Learning -- Chapter 3: Deep Learning and Neutral Networks Overview -- Chapter 4: Complimentary Libraries to TensorFlow 2.x -- Chapter 5: A Guide to TensorFlow 2.0 and Deep Learning Pipeline -- Chapter 6: Feedfoward Neutral Networks -- Chapter 7: Convolutional Neural Networks -- Chapter 8: Recurrent Neural Networks -- Chapter 9: Natural Language Processing -- Chapter 10: Recommender Systems -- Chapter 11: Auto-Encoders -- Chapter 12: Generative Adversarial Networks. Requires an SPL library card. Implement deep learning applications using TensorFlow while learning the "why" through in-depth conceptual explanations. You'll start by learning what deep learning offers over other machine learning models. Then familiarize yourself with several technologies used to create deep learning models. While some of these technologies are complementary, such as Pandas, Scikit-Learn, and Numpy-others are competitors, such as PyTorch, Caffe, and Theano. This book clarifies the positions of deep learning and Tensorflow among their peers. You'll then work on supervised deep learning models to gain applied experience with the technology. A single-layer of multiple perceptrons will be used to build a shallow neural network before turning it into a deep neural network. After showing the structure of the ANNs, a real-life application will be created with Tensorflow 2.0 Keras API. Next, you'll work on data augmentation and batch normalization methods. Then, the Fashion MNIST dataset will be used to train a CNN. CIFAR10 and Imagenet pre-trained models will be loaded to create already advanced CNNs. Finally, move into theoretical applications and unsupervised learning with auto-encoders and reinforcement learning with tf-agent models. With this book, you'll delve into applied deep learning practical functions and build a wealth of knowledge about how to use TensorFlow effectively. You will: Compare competing technologies and see why TensorFlow is more popular Generate text, image, or sound with GANs Predict the rating or preference a user will give to an item Sequence data with recurrent neural networks. Mode of access: World Wide Web. Artificial intelligence 31959 Machine Learning 31960 Deep Learning and Neural networks 31961 Complementary Libraries to tensor flow 31962 Electronic books. local 31963 SpringerLink (Online service) 31964 O'Reilly (Firm) 31965 Serials Solutions 31966 Springer Nature eBook Printed edition: 9781484265123 Printed edition: 9781484265147 View this electronic item in O'Reilly Online Learning: Academic/Public Library Edition. https://ezproxy.spl.org/login?url=https://learning.oreilly.com/library/view/~/9781484265130/?ar An e-book available through full-text database. ddc BK 1st 006.32 YALO 222637 222637 03678066 0 0 ddc 0 0 MCA AIMIT AIMIT 2022-03-24 Biblios Book Point 899.00 Bill no:6623; Bill dt:2022-03-23 0 006.32 YALO MCA17034 2026-05-23 00:00:00 719.20 2022-04-26 BK