The tensorflow workshop : a hands-on guide to building deep learning models from scratch using real-world datasets / Matthew Moocarme, Anthony So and Anthony Maddalone.

Format:

Book

Author/Creator:

Moocarme, Matthew, author.

So, Anthony Veasna, 1992-2020, author.

Maddalone, Anthony, author.

Language:

English

Subjects (All):

TensorFlow.

Machine learning.

Neural networks (Computer science).

Physical Description:

1 online resource (601 pages)

Edition:

1st edition.

Place of Publication:

Birmingham, England ; Mumbai : Packt, [2021]

Biography/History:

Moocarme Matthew: Matthew Moocarme is an accomplished data scientist with more than eight years of experience in creating and utilizing machine learning models. He comes from a background in the physical sciences, in which he holds a Ph. D. in physics from the Graduate Center of CUNY. Currently, he leads a team of data scientists and engineers in the media and advertising space to build and integrate machine learning models for a variety of applications. In his spare time, Matthew enjoys sharing his knowledge with the data science community through published works, conference presentations, and workshops. Bagchi Abhranshu: Contacted by Sneha Shinde on Jan 28, 2020So Anthony: Anthony So is a renowned leader in data science. He has extensive experience in solving complex business problems using advanced analytics and AI in different industries including financial services, media, and telecommunications. He is currently the chief data officer of one of the most innovative fintech start-ups. He is also the author of several best-selling books on data science, machine learning, and deep learning. He has won multiple prizes at several hackathon competitions, such as Unearthed, GovHack, and Pepper Money. Anthony holds two master's degrees, one in computer science and the other in data science and innovation. Maddalone Anthony: Anthony Maddalone is a research engineer at TieSet, a Silicon Valley-based leader in distributed artificial intelligence and federated learning. He is a former founder and CEO of a successful start-up. Anthony lives with his wife and two children in Colorado, where they enjoy spending time outdoors. He is also a master's candidate in analytics with a focus on industrial engineering at the Georgia Institute of Technology.

Summary:

This Workshop will teach you how to build deep learning models from scratch using real-world datasets with the TensorFlow framework. You will gain the knowledge you need to process a variety of data types, perform tensor computations, and understand the different layers in a deep learning model.

Contents:

Cover

FM

Copyright

Table of Contents

Preface

Chapter 1: Introduction to Machine Learning with TensorFlow

Introduction

Implementing Artificial Neural Networks in TensorFlow

Advantages of TensorFlow

Disadvantages of TensorFlow

The TensorFlow Library in Python

Exercise 1.01: Verifying Your Version of TensorFlow

Introduction to Tensors

Scalars, Vectors, Matrices, and Tensors

Exercise 1.02: Creating Scalars, Vectors, Matrices, and Tensors in TensorFlow

Tensor Addition

Exercise 1.03: Performing Tensor Addition in TensorFlow

Activity 1.01: Performing Tensor Addition in TensorFlow

Reshaping

Tensor Transposition

Exercise 1.04: Performing Tensor Reshaping and Transposition in TensorFlow

Activity 1.02: Performing Tensor Reshaping and Transposition in TensorFlow

Tensor Multiplication

Exercise 1.05: Performing Tensor Multiplication in TensorFlow

Optimization

Forward Propagation

Backpropagation

Learning Optimal Parameters

Optimizers in TensorFlow

Activation functions

Activity 1.03: Applying Activation Functions

Summary

Chapter 2: Loading and Processing Data

Exploring Data Types

Data Preprocessing

Processing Tabular Data

Exercise 2.01: Loading Tabular Data and Rescaling Numerical Fields

Activity 2.01: Loading Tabular Data and Rescaling Numerical Fields with a MinMax Scaler

Exercise 2.02: Preprocessing Non-Numerical Data

Processing Image Data

Exercise 2.03: Loading Image Data for Batch Processing

Image Augmentation

Activity 2.02: Loading Image Data for Batch Processing

Text Processing

Exercise 2.04: Loading Text Data for TensorFlow Models

Audio Processing

Exercise 2.05: Loading Audio Data for TensorFlow Models

Activity 2.03: Loading Audio Data for Batch Processing

Summary.

Chapter 3: TensorFlow Development

TensorBoard

Exercise 3.01: Using TensorBoard to Visualize Matrix Multiplication

Activity 3.01: Using TensorBoard to Visualize Tensor Transformations

Exercise 3.02: Using TensorBoard to Visualize Image Batches

TensorFlow Hub

Exercise 3.03: Downloading a Model from TensorFlow Hub

Google Colab

Advantages of Google Colab

Disadvantages of Google Colab

Development on Google Colab

Exercise 3.04: Using Google Colab to Visualize Data

Activity 3.02: Performing Word Embedding from a Pre-Trained Model from TensorFlow Hub

Chapter 4: Regression and Classification Models

Sequential Models

Keras Layers

Exercise 4.01: Creating an ANN with TensorFlow

Model Fitting

The Loss Function

Model Evaluation

Exercise 4.02: Creating a Linear Regression Model as an ANN with TensorFlow

Exercise 4.03: Creating a Multi-Layer ANN with TensorFlow

Activity 4.01: Creating a Multi-Layer ANN with TensorFlow

Classification Models

Exercise 4.04: Creating a Logistic Regression Model as an ANN with TensorFlow

Activity 4.02: Creating a Multi-Layer Classification ANN with TensorFlow

Chapter 5: Classification Models

Binary Classification

Logistic Regression

Binary Cross-Entropy

Binary Classification Architecture

Exercise 5.01: Building a Logistic Regression Model

Metrics for Classifiers

Accuracy and Null Accuracy

Precision, Recall, and the F1 Score

Confusion Matrices

Exercise 5.02: Classification Evaluation Metrics

Multi-Class Classification

The Softmax Function

Categorical Cross-Entropy

Multi-Class Classification Architecture

Exercise 5.03: Building a Multi-Class Model

Activity 5.01: Building a Character Recognition Model with TensorFlow

Multi-Label Classification.

Activity 5.02: Building a Movie Genre Tagging a Model with TensorFlow

Chapter 6: Regularization and Hyperparameter Tuning

Regularization Techniques

L1 Regularization

L2 Regularization

Exercise 6.01: Predicting a Connect-4 Game Outcome Using the L2 Regularizer

Dropout Regularization

Exercise 6.02: Predicting a Connect-4 Game Outcome Using Dropout

Early Stopping

Activity 6.01: Predicting Income with L1 and L2 Regularizers

Hyperparameter Tuning

Keras Tuner

Random Search

Exercise 6.03: Predicting a Connect-4 Game Outcome Using Random Search from Keras Tuner

Hyperband

Exercise 6.04: Predicting a Connect-4 Game Outcome Using Hyperband from Keras Tuner

Bayesian Optimization

Activity 6.02: Predicting Income with Bayesian Optimization from Keras Tuner

Chapter 7: Convolutional Neural Networks

CNNs

Image Representation

The Convolutional Layer

Creating the Model

Exercise 7.01: Creating the First Layer to Build a CNN

Pooling Layer

Max Pooling

Average Pooling

Exercise 7.02: Creating a Pooling Layer for a CNN

Flattening Layer

Exercise 7.03: Building a CNN

Batch Normalization

Exercise 7.04: Building a CNN with Additional Convolutional Layers

Binary Image Classification

Object Classification

Exercise 7.05: Building a CNN

Activity 7.01: Building a CNN with More ANN Layers

Chapter 8: Pre-Trained Networks

ImageNet

Transfer Learning

Exercise 8.01: Classifying Cats and Dogs with Transfer Learning

Fine-Tuning

Activity 8.01: Fruit Classification with Fine-Tuning

Feature Extraction

Activity 8.02: Transfer Learning with TensorFlow Hub

Chapter 9: Recurrent Neural Networks

Sequential Data.

Examples of Sequential Data

Exercise 9.01: Training an ANN for Sequential Data - Nvidia Stock Prediction

Recurrent Neural Networks

RNN Architecture

Vanishing Gradient Problem

Long Short-Term Memory Network

Exercise 9.02: Building an RNN with an LSTM Layer - Nvidia Stock Prediction

Activity 9.01: Building an RNN with Multiple LSTM Layers to Predict Power Consumption

Natural Language Processing

Dataset Cleaning

Generating a Sequence and Tokenization

Padding Sequences

Back Propagation Through Time (BPTT)

Exercise 9.03: Building an RNN with an LSTM Layer for Natural Language Processing

Activity 9.02: Building an RNN for Predicting Tweets' Sentiment

Chapter 10: Custom TensorFlow Components

TensorFlow APIs

Implementing Custom Loss Functions

Building a Custom Loss Function with the Functional API

Building a Custom Loss Function with the Subclassing API

Exercise 10.01: Building a Custom Loss Function

Implementing Custom Layers

Introduction to ResNet Blocks

Building Custom Layers with the Functional API

Building Custom Layers with Subclassing

Exercise 10.02: Building a Custom Layer

Activity 10.01: Building a Model with Custom Layers and a Custom Loss Function

Chapter 11: Generative Models

Text Generation

Extending NLP Sequence Models to Generate Text

Generating a Sequence of n-gram Tokens

Exercise 11.01: Generating Text

Generative Adversarial Networks

The Generator Network

The Discriminator Network

The Adversarial Network

Combining the Generative and Discriminative Models

Generating Real Samples with Class Labels

Creating Latent Points for the Generator.

Using the Generator to Generate Fake Samples and Class Labels

Evaluating the Discriminator Model

Training the Generator and Discriminator

Creating the Latent Space, Generator, Discriminator, GAN, and Training Data

Exercise 11.02: Generating Sequences with GANs

Deep Convolutional Generative Adversarial Networks (DCGANs)

Training a DCGAN

Exercise 11.03: Generating Images with DCGAN

Activity 11.01: Generating Images Using GANs

Appendix

Index.

Notes:

Description based on print version record.

ISBN:

1-80020-022-6

OCLC:

1290491271

Publisher Number:

9781800205253