Lavoisier S.A.S.
14 rue de Provigny
94236 Cachan cedex
FRANCE

Heures d'ouverture 08h30-12h30/13h30-17h30
Tél.: +33 (0)1 47 40 67 00
Fax: +33 (0)1 47 40 67 02


Url canonique : www.lavoisier.fr/livre/autre/deep-learning-pipeline/descriptif_4207408
Url courte ou permalien : www.lavoisier.fr/livre/notice.asp?ouvrage=4207408

Deep Learning Pipeline, 1st ed. Building a Deep Learning Model with TensorFlow

Langue : Anglais

Auteurs :

Couverture de l’ouvrage Deep Learning Pipeline

Build your own pipeline based on modern TensorFlow approaches rather than outdated engineering concepts. This book shows you how to build a deep learning pipeline for real-life TensorFlow projects. 

You'll learn what a pipeline is and how it works so you can build a full application easily and rapidly. Then troubleshoot and overcome basic Tensorflow obstacles to easily create functional apps and deploy well-trained models. Step-by-step and example-oriented instructions help you understand each step of the deep learning pipeline while you apply the most straightforward and effective tools to demonstrative problems and datasets.  

You'll also develop a deep learning project by preparing data, choosing the model that fits that data, and debugging your model to get the best fit to data all using Tensorflow techniques. Enhance your skills by accessing some of the most powerful recent trends in data science. If you've ever considered building your own image or text-tagging solution or entering a Kaggle contest, Deep Learning Pipeline is for you!

What You'll Learn
  • Develop a deep learning project using data
  • Study and apply various models to your data
  • Debug and troubleshoot the proper model suited for your data

Who This Book Is For

Developers, analysts, and data scientists looking to add to or enhance their existing skills by accessing some of the most powerful recent trends in data science. Prior experience in Python or other TensorFlow related languages and mathematics would be helpful.

Part One:​ Introduction
Prepares the readers with the prerequisites needed. 

 

Chapter 1:​ Tools, Theories, and Equations
This chapter provides the big picture that shows the audience the field that the book describes. Introduces the mathematical equations and notations that describe how machine ;earning works, the programming tools and packages needed in this book, and some theories. 

  • Probability Theory, Decision Theory and Information Theory 
  • Introduction to machine learning 

o What is machine learning 

o What is deep learning 

  • From machine learning to deep learning 
  • Mathematical notation 
  • Python installation 

o Python and Jupyter
oCommon Deep-Learning Packages 
oTensorFlow Installation 

·      Summary

·        

Chapter 2:​ A Tour Through the Deep Learning Pipeline
In chapter two, we introduce the pipeline. What are the deep learning approaches and related sub-fields. What are the steps of a deep learning pipeline. And what are the extras added to TensorFlow that make it unique compared to other deep learning frameworks. 

● Deep Learning Approaches 

● Deep Learning Pipeline 

o Data
oGoals
oModels
oFeatures
oModel Evaluation 

● Fast preview of the TensorFlow pipeline 

● Summary 

 

Chapter 3: ​Build Your First Toy TensorFlow App
To make sure that we don’t drop the audience into the middle things without setup, we will show them a small example using TensorFlow that quickly introduces each step of the deep learning pipeline. And make sure that the audience knows each step of the pipeline, how it is important, and how to use it. 

  • TensorFlow Basics for Development 
  • XOR Implementation Using TensorFlow 
  • Linear Regression in TensorFlow 
  • Summary

Part Two: ​Data
Covers everything about data. From data collection to understanding intuition to data processing and preparation. 

 

Chapter 4:​ Defining Data
This chapter as its name suggests is about defining data. Readers should know the type of data they are dealing with so they can choose the right approach for preparing that data. 

  • Defining Data 
  • Why should you read this chapter? 
  • Structured, semi-structured, and unstructured data 
  • Divide and Conquer 
  • The types of data you will deal with 

o Tabular (Numerical and Categorical) 

    • Quantitative versus Qualitative data 
    •  The four levels of data 
      • Nominal level 
      • Ordinal level 
      • Interval level 
      • Ratio level 
    • Example – Titanic 

o Text 

    • Example – Classifying IMDB Movie Reviews o Images
    • Type of images (2D, 3D, 4D) 
    • Example – CIFAR-10 

·      Quick recap and check 

·      Summary 

 

Chapter 5:​ Data Wrangling and Preprocessing
After understanding the data, readers now choose the approaches and methodologies for preparing it.

  • The deep learning pipeline revisited 
  • Data loading and preprocessing 

o Data Loading with Numpy 

o Data Loading with Pandas 

  • Missing and Noisy Data 
  • Dealing with big datasets 
  • Accessing other data formats 
  • Data preprocessing 
  • Data augmentation 
  • Categorical to Text data 
  • Methods can not use in text data 
  • Tokenization, Steaming, and Stop-words 
  • The Bag-of-Words Model (BoW) 

o CountVectorizer and Term Frequency-Inverse Document Frequency (TF-IDF) 

  • Summary

 

Chapter 6:​ Data Resampling
After cleaning and preparing the dataset, the reader should know how to sample this dataset in the right way. Choosing wrong samples from your data may influence the result of your models. So in this chapter, we will illustrate all techniques and approaches needed to sample your dataset in the right way. 

  • Creating training and test sets 
  • Cross-Validation

o Validation set technique
oLeave-One-Out Cross-Validation (LOOCV) o K-Fold Cross-Validation 

  • Bootstrap
  • Case Study -- Generators (modify) 
  • Summary

 

Chapter 7:​ Feature Selection and Feature Engineering
In this chapter, we will describe a very important topic in the pipeline, which is feature selection and engineering. The audience should know how to select and choose the most important input feature that contributes most to your output feature. We will cover the process of using domain knowledge of the data to create features that make machine learning algorithms work. And also feature selection and engineering, which are fundamental to the application of machine and deep learning. 

  • Datasets used in this chapter 
  • Feature selection and filtering 
  • Principal component analysis 
  • o Non-Negative Matrix Factorization (NMF) o Sparse PCA
    oKernel PCA 

    • ZCA Whitening 
    • Topic Model and Latent Dirichlet Allocation (LDA) 
    • Atom extraction and dictionary learning 
    • Summary

     

    Part Three: ​TensorFlow
    In this part, first we illustrate the fundamental concepts of deep learning. Then we deep dive to the core of neural networks and the types of neural networks, describing each type and showing the important concepts of the equations of deep learning. Finally we present real-life examples of each type. 

     

    Chapter 8:​ Deep Learning Fundamentals
    In this chapter, we will describe the basic functions that deep learning is built on. Then we will try to build layers from these functions, and combine these layers together to get more complex models that will help us solve more complex problems. All that will be described by TensorFlow examples. 

    • Perceptron
      oSingle Perceptron 

    o Multi Perceptron 

    • Difference neural network layers 

    o input layer 

    o hidden layer

    o output layer

    ▪sigmoid

    ▪Softmax

    • Shallow vs. Deep Neural Networks 
    • Activation Functions 
    • Gradient Descent 
    • Loss Functions and Backpropagation 
    • Vanishing and Exploding Gradients 
    • TensorFlow Basics 

    o Placeholder vs. Variable vs. Constant
    oGradient-Descent Optimization Methods from a Deep-Learning Perspective 

    o Learning Rate in Mini-batch Approach to Stochastic Gradient Descent
    oOptimizers in TensorFlow 

    • Full-Batch vs. Stochastic vs. Mini-Batch Gradient Descent 
    • Summary

     

    Chapter 9:​ Improving Deep Neural Networks
    After building the deep learning model, the improvement starts. This chapter covers optimization, tuning, and choosing hyperparameter techniques. The reader will also learn to evaluate, optimize, and tune their models parameters to reach optimal solutions and accuracy. 

    • Gradient-Descent Optimization Methods 
    • Choosing the Learning Rate 
    • Dropout Layers and Regularization 
    • Optimizers in TensorFlow 
    • Normalization Techniques 

    o Batch Normalization 

    o Weight Normalization 

    o Group Normalization
    oInstance Normalization 

    o Layer Normalization 

    • ●  Summary

     

    Chapter 10:​ Convolutional Neural Networks
    One of the important classes of deep learning is the convolutional neural network. In this chapter, we will illustrate everything about CNN from the one-dimensional mask to the advanced stuff like weight sharing and the difference between equivariance and invariance. We will illustrate a case study using the famous dataset CIFAR-10 

    ● What is ConvNet 

    • Convolution Operation 

    o 1D Convolution 

    o 2D Convolution 

    • Two-dimensional Unit Step Function 
    • Common Image-Processing Filters 

    o Mean and Median Filters
    oGaussian Filter
    oGradient-based Filters
    oSobel Edge-Detection Filter o Identity Transform 

    • Convolution Neural Networks 
    • Layers of Convolution Neural Networks 

    o Input Layer
    oConvolution Layer oPooling Layer 

      Backpropagation Through the Convolutional and Pooling Layer 
    • Weight Sharing Through Convolution and Its Advantages 
    • Translation Equivariance and Invariance 
    • Case Study -- Digit Recognition on the CIFAR-10 Dataset 
    • Summary

     

    Chapter 11:​ Sequential Models
    Another class of deep learning is the sequential models. In this chapter, we will describe the problem of sequential data and the rise of recurrent neural networks as well as the problem and also the evolution of the GRU and LSTM. 

    • Recurrent Neural Networks (RNN) 
    • Language Modeling 
    • Backpropagation Through Time (BPTT) 
    • Case Study -- Predicting the Next Word in a Sentence 
    • Vanishing and Exploding Gradient Problem in RNN 
    • Solution to Vanishing and Exploding Gradients Problem in RNNs 
    • Long Short-Term Memory (LSTM) 
    • Exploding and Vanishing Gradient Reduction in LSTM 
    • Case Study -- Digit Identification on the MNIST Dataset 
    • Gated Recurrent Unit (GRU) 
    • Bidirectional RNN (Bi-RNN) 
    • Summary

     

    Part Four:​ Applications and Appendix
    We will give the readers this part to make sure that they practice by using TensorFlow and build the pipeline. 

     

    Chapter 12:​ Selected Topics in Computer Vision
    After finishing CNN in part three, it’s good to add some extra knowledge that makes it easier for readers when they work, like using pre-built architectures and transfer learning. 

    • ●  Different Architectures in Convolutional Neural Networks 

    o LeNet 

    o AlexNet 

    o VGG16 

    o ResNet 

    • Transfer Learning 
    • Guidelines for Using Transfer Learning 

    o InceptionV3 

    o Pre-trained VGG16 

    ● Summary 

     

    Chapter 13:​ Selected Topics in Natural Language Processing
    In this chapter, we fill the gaps that the audience needs for working with text. Giving the readers all the advanced approaches and techniques of natural language processing. 

    • Vector Space Model 
    • Vector Representation of Words 
    • Word2Vec
    o Continuous Bag of Words (CBOW)
    oSkip-Gram Model for Word Embedding 

    o GloVe 

    ● Summary 

     

    Chapter 14:​ Model Deployment
    Here we will show the audience how model deployment works. It’s one last important step needed to communicate with others such as stakeholders and decision makers. 

    • Introduction to Model Deployment. 
    • Type and Architecture of deployments. 
    • Introduction to Flask. 
      • Installing Flask 
      • Building a small REST Service 
    • Real-life train and test. 
    • Case Study -- tiny predictors 
    • Summary

     

    Chapter 15:​ Applications
    Here we will show some case studies to make sure that the reader gets the full knowledge and understanding of how to build pipelines with real-life examples. 

    • Case Study -- Tabular Datasets 
    • Case Study -- IMDB Movie Review Data with Paragraph2Vec 
    • Case Study -- Image Segmentation 
    • Conclusion

    Hisham Elamir​ is a data scientist with expertise in machine learning, deep learning, and statistics. He currently lives and works in Cairo, Egypt. In his work projects, he faces challenges ranging from natural language processing (NLP), behavioral analysis, and machine learning to distributed processing. He is very passionate about his job and always tries to stay updated about the latest developments in data science technologies, attending meetups, conferences, and other events. 

     Mahmoud Hamdy is a machine learning engineer who works in Egypt and lives in Egypt, His primary area of study is the overlap between knowledge, logic, language, and learning. He works helping train machine learning, and deep learning models to distil large amounts of unstructured, semi-structured, and structured data into new knowledge about the world by using methods ranging from deep learning to statistical relational learning. He applies strong theoretical and practical skills in several areas of machine learning to finding novel and effective solutions for interesting and challenging problems in such interconnections

    Discover the difference between the regular and pipelined model of deep learning

    Learn all the detailed steps of an applicable deep learning pipeline

    Use a pipeline to help better manage deep learning projects and applications

    Date de parution :

    Ouvrage de 551 p.

    15.5x23.5 cm

    Disponible chez l'éditeur (délai d'approvisionnement : 15 jours).

    47,46 €

    Ajouter au panier