| Document Type
|
:
|
BL
|
| Record Number
|
:
|
851306
|
| Main Entry
|
:
|
Michelucci, Umberto
|
| Title & Author
|
:
|
Advanced applied deep learning : : convolutional neural networks and object detection /\ Umberto Michelucci.
|
| Publication Statement
|
:
|
New York :: Apress,, [2019]
|
|
|
:
|
, ©2019
|
| Page. NO
|
:
|
1 online resource :: illustrations (some color)
|
| ISBN
|
:
|
1484249755
|
|
|
:
|
: 1484249763
|
|
|
:
|
: 1484249771
|
|
|
:
|
: 9781484249758
|
|
|
:
|
: 9781484249765
|
|
|
:
|
: 9781484249772
|
|
|
:
|
9781484249758
|
| Bibliographies/Indexes
|
:
|
Includes bibliographical references and index.
|
| Contents
|
:
|
Intro; Table of Contents; About the Author; About the Technical Reviewer; Acknowledgments; Introduction; Chapter 1: Introduction and Development Environment Setup; GitHub Repository and Companion Website; Mathematical Level Required; Python Development Environment; Google Colab; Benefits and Drawbacks to Google Colab; Anaconda; Installing TensorFlow the Anaconda Way; Local Jupyter Notebooks; Benefits and Drawbacks to Anaconda; Docker Image; Benefits and Drawbacks to a Docker Image; Which Option Should You Choose?; Chapter 2: TensorFlow: Advanced Topics; Tensorflow Eager Execution
|
|
|
:
|
Chapter 5: Cost Functions and Style TransferComponents of a Neural Network Model; Training Seen as an Optimization Problem; A Concrete Example: Linear Regression; The Cost Function; Mathematical Notation; Typical Cost Functions; Mean Square Error; Intuitive Explanation; MSE as the Second Moment of a Moment-Generating Function; Cross-Entropy; Self-Information or Suprisal of an Event; Suprisal Associated with an Event X; Cross-Entropy; Cross-Entropy for Binary Classification; Cost Functions: A Final Word; Neural Style Transfer; The Mathematics Behind NST; An Example of Style Transfer in Keras
|
|
|
:
|
Dataset AbstractionIterating Over a Dataset; Simple Batching; Simple Batching with the MNIST Dataset; Using tf.data. Dataset in Eager Execution Mode; Conclusions; Chapter 3: Fundamentals of Convolutional Neural Networks; Kernels and Filters; Convolution; Examples of Convolution; Pooling; Padding; Building Blocks of a CNN; Convolutional Layers; Pooling Layers; Stacking Layers Together; Number of Weights in a CNN; Convolutional Layer; Pooling Layer; Dense Layer; Example of a CNN: MNIST Dataset; Visualization of CNN Learning; Brief Digression: keras.backend.function(); Effect of Kernels
|
|
|
:
|
Effect of Max-PoolingChapter 4: Advanced CNNs and Transfer Learning; Convolution with Multiple Channels; History and Basics of Inception Networks; Inception Module: Naïve Version; Number of Parameters in the Naïve Inception Module; Inception Module with Dimension Reduction; Multiple Cost Functions: GoogLeNet; Example of Inception Modules in Keras; Digression: Custom Losses in Keras; How To Use Pre-Trained Networks; Transfer Learning: An Introduction; A Dog and Cat Problem; Classical Approach to Transfer Learning; Experimentation with Transfer Learning
|
|
|
:
|
Enabling Eager ExecutionPolynomial Fitting with Eager Execution; MNIST Classification with Eager Execution; TensorFlow and Numpy Compatibility; Hardware Acceleration; Checking the Availability of the GPU; Device Names; Explicit Device Placement; GPU Acceleration Demonstration: Matrix Multiplication; Effect of GPU Acceleration on the MNIST Example; Training Only Specific Layers; Training Only Specific Layers: An Example; Removing Layers; Keras Callback Functions; Custom Callback Class; Example of a Custom Callback Class; Save and Load Models; Save Your Weights Manually; Saving the Entire Model
|
| Abstract
|
:
|
Develop and optimize deep learning models with advanced architectures. This book teaches you the intricate details and subtleties of the algorithms that are at the core of convolutional neural networks. In Advanced Applied Deep Learning, you will study advanced topics on CNN and object detection using Keras and TensorFlow. Along the way, you will look at the fundamental operations in CNN, such as convolution and pooling, and then look at more advanced architectures such as inception networks, resnets, and many more. While the book discusses theoretical topics, you will discover how to work efficiently with Keras with many tricks and tips, including how to customize logging in Keras with custom callback classes, what is eager execution, and how to use it in your models. Finally, you will study how object detection works, and build a complete implementation of the YOLO (you only look once) algorithm in Keras and TensorFlow. By the end of the book you will have implemented various models in Keras and learned many advanced tricks that will bring your skills to the next level. You will: See how convolutional neural networks and object detection work Save weights and models on disk Pause training and restart it at a later stage Use hardware acceleration (GPUs) in your code Work with the Dataset TensorFlow abstraction and use pre-trained models and transfer learning Remove and add layers to pre-trained networks to adapt them to your specific project Apply pre-trained models such as Alexnet and VGG16 to new datasets.
|
| Subject
|
:
|
Machine learning.
|
| Subject
|
:
|
Neural networks (Computer science)
|
| Subject
|
:
|
Python (Computer program language)
|
| Subject
|
:
|
Machine learning.
|
| Subject
|
:
|
Neural networks (Computer science)
|
| Subject
|
:
|
Python (Computer program language)
|
| Dewey Classification
|
:
|
006.3/1
|
| LC Classification
|
:
|
Q325.5.M53 2019eb
|