Manaswi Deep Learning with Applications Using Python

Chatbots and Face, Object, and Speech Recognition With TensorFlow and Keras
1. ed
ISBN: 978-1-4842-3516-4
Verlag: Apress
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 228 Seiten

Build deep learning applications, such as computer vision, speech recognition, and chatbots, using frameworks such as TensorFlow and Keras. This book helps you to ramp up your practical know-how in a short period of time and focuses you on the domain, models, and algorithms required for deep learning applications. Deep Learning with Applications Using Python covers topics such as chatbots, natural language processing, and face and object recognition. The goal is to equip you with the concepts, techniques, and algorithm implementations needed to create programs capable of performing deep learning.
This book covers intermediate and advanced levels of deep learning, including convolutional neural networks, recurrent neural networks, and multilayer perceptrons. It also discusses popular APIs such as IBM Watson, Microsoft Azure, and scikit-learn. 
What You Will Learn Work with various deep learning frameworks such as TensorFlow, Keras, and scikit-learn.
Build face recognition and face detection capabilities
Create speech-to-text and text-to-speech functionality
Make chatbots using deep learning

Who This Book Is For
Data scientists and developers who want to adapt and build deep learning applications.



Navin K Manaswi has been developing AI solutions/products with the use of cutting edge technologies and sciences related to Artificial Intelligence for many years. Having worked for Consulting companies in Malaysia, Singapore and Dubai Smart City project, he has developed a rare skill of delivering end-to-end data science solutions. He has been building solutions for video intelligence, document intelligence and human-like chatbots in his own company. Through this book, he wants to democratize the cognitive computing and services for everyone specially developers, data scientists, software engineers, database engineers, data analysts and CXOs.

Autoren/Hrsg.

Manaswi, Navin Kumar

Inhaltsverzeichnis

1;Table of Contents;4
2;Foreword by Tarry Singh;10
3;About the Author;13
4;About the Technical Reviewer;14
5;Chapter 1: Basics of TensorFlow;15
5.1;Tensors;16
5.2;Computational Graph and Session;17
5.3;Constants, Placeholders, and Variables;20
5.4;Placeholders;23
5.5;Creating Tensors;26
5.5.1;Fixed Tensors;27
5.5.2;Sequence Tensors;28
5.5.3;Random Tensors;29
5.6;Working on Matrices;30
5.7;Activation Functions;31
5.7.1;Tangent Hyperbolic and Sigmoid;32
5.7.2;ReLU and ELU;33
5.7.3;ReLU6;34
5.8;Loss Functions;36
5.8.1;Loss Function Examples;37
5.8.2;Common Loss Functions;37
5.9;Optimizers;39
5.9.1;Loss Function Examples;40
5.9.2;Common Optimizers;41
5.10;Metrics;42
5.10.1;Metrics Examples;42
5.10.2;Common Metrics;43
6;Chapter 2: Understanding and  Working with Keras;45
6.1;Major Steps to Deep Learning Models;46
6.1.1;Load Data;47
6.1.2;Preprocess the Data;47
6.1.3;Define the Model;48
6.1.4;Compile the Model;50
6.1.5;Fit the Model;51
6.1.6;Evaluate Model;52
6.1.7;Prediction;52
6.1.8;Save and Reload the Model;53
6.1.9;Optional: Summarize the Model;53
6.2;Additional Steps to Improve Keras Models;54
6.3;Keras with TensorFlow;56
7;Chapter 3: Multilayer Perceptron;58
7.1;Artificial Neural Network;58
7.2;Single-Layer Perceptron;60
7.3;Multilayer Perceptron;60
7.4;Logistic Regression Model;62
8;Chapter 4: Regression to MLP in TensorFlow;70
8.1;TensorFlow Steps to Build Models;70
8.2;Linear Regression in TensorFlow;71
8.3;Logistic Regression Model;75
8.4;Multilayer Perceptron in TensorFlow;78
9;Chapter 5: Regression to MLP in Keras;82
9.1;Log-Linear Model;82
9.2;Keras Neural Network for Linear Regression;84
9.3;Logistic Regression;86
9.3.1;scikit-learn for Logistic Regression;87
9.3.2;Keras Neural Network for Logistic Regression;87
9.3.3;Fashion MNIST Data: Logistic Regression in Keras;90
9.4;MLPs on the Iris Data;93
9.4.1;Write the Code;93
9.4.2;Build a Sequential Keras Model;94
9.5;MLPs on MNIST Data (Digit Classification);97
9.6;MLPs on Randomly Generated Data;101
10;Chapter 6: Convolutional Neural Networks;103
10.1;Different Layers in a CNN;103
10.2;CNN Architectures;107
11;Chapter 7: CNN in TensorFlow;109
11.1;Why TensorFlow for CNN Models?;109
11.2;TensorFlow Code for Building an Image Classifier for MNIST Data;110
11.3;Using a High-Level API for Building CNN Models;116
12;Chapter 8: CNN in Keras;117
12.1;Building an Image Classifier for MNIST Data in Keras;117
12.1.1;Define the Network Structure;119
12.1.2;Define the Model Architecture;120
12.2;Building an Image Classifier with CIFAR-10 Data;122
12.2.1;Define the Network Structure;123
12.3;Define the Model Architecture;124
12.4;Pretrained Models;125
13;Chapter 9: RNN and LSTM;127
13.1;The Concept of RNNs;127
13.2;The Concept of LSTM;130
13.3;Modes of LSTM;130
13.4;Sequence Prediction;131
13.4.1;Sequence Numeric Prediction;132
13.4.2;Sequence Classification;132
13.4.3;Sequence Generation;133
13.4.4;Sequence-to-Sequence Prediction;133
13.5;Time-Series Forecasting with the LSTM Model;134
14;Chapter 10: Speech to Text and Vice Versa;139
14.1;Speech-to-Text Conversion;140
14.2;Speech as Data;140
14.3;Speech Features: Mapping Speech to a Matrix;141
14.4;Spectrograms: Mapping Speech to an Image;143
14.5;Building a Classifier for Speech Recognition Through MFCC Features;144
14.6;Building a Classifier for Speech Recognition Through a Spectrogram;145
14.7;Open Source Approaches;147
14.8;Examples Using Each API;147
14.8.1;Using PocketSphinx;147
14.8.2;Using the Google Speech API;148
14.8.3;Using the Google Cloud Speech API;149
14.8.4;Using the Wit.ai API;149
14.8.5;Using the Houndify API;150
14.8.6;Using the IBM Speech to Text API;150
14.8.7;Using the Bing Voice Recognition API;151
14.9;Text-to-Speech Conversion;152
14.9.1;Using pyttsx;152
14.9.2;Using SAPI;152
14.9.3;Using SpeechLib;152
14.9.4;Audio Cutting Code;153
14.10;Cognitive Service Providers;154
14.10.1;Microsoft Azure;155
14.10.2;Amazon Cognitive Services;155
14.10.3;IBM Watson Services;156
14.11;The Future of Speech Analytics;156
15;Chapter 11: Developing Chatbots;157
15.1;Why Chatbots?;158
15.2;Designs and Functions of Chatbots;158
15.3;Steps for Building a Chatbot;159
15.3.1;Preprocessing Text and Messages;160
15.3.1.1;Tokenization;160
15.3.1.2;Removing Punctuation Marks;160
15.3.1.3;Removing Stop Words;161
15.3.1.4;Named Entity Recognition;162
15.3.1.4.1;Using Stanford NER;162
15.3.1.4.2;Using MITIE NER (Pretrained);163
15.3.1.4.3;Using MITIE NER (Self-Trained);163
15.3.1.5;Intent Classification;164
15.3.1.5.1;Word Embedding;165
15.3.1.5.1.1;Count Vector;166
15.3.1.5.1.2;Term Frequency-Inverse Document Frequency (TF-IDF);166
15.3.1.5.2;Word2Vec;169
15.3.1.6;Building the Response;177
15.3.2;Chatbot Development Using APIs;178
15.3.2.1;Cognitive Services of Microsoft Azure;179
15.3.2.2;Amazon Lex;180
15.3.2.3;IBM Watson;180
15.4;Best Practices of Chatbot Development;181
15.4.1;Know the Potential Users;181
15.4.2;Read the User Sentiments and Make the Bot Emotionally Enriching;181
16;Chapter 12: Face Detection and Recognition;183
16.1;Face Detection, Face Recognition, and Face Analysis;184
16.2;OpenCV;184
16.2.1;Eigenfaces;185
16.2.2;LBPH;187
16.2.3;Fisherfaces;188
16.3;Detecting a Face;189
16.4;Tracking the Face;191
16.5;Face Recognition;194
16.6;Deep Learning–Based Face Recognition;197
16.7;Transfer Learning;200
16.7.1;Why Transfer Learning?;200
16.7.2;Transfer Learning Example;201
16.7.3;Calculate the Transfer Value;203
16.8;APIs;209
17;Appendix 1: Keras Functions for Image Processing;212
18;Appendix 2: Some of the Top Image Data Sets Available;217
19;Appendix 3: Medical Imaging: DICOM File Format;220
19.1;Why DICOM?;220
19.2;What Is the DICOM File Format?;220
20;Index;222