Cover image for Learn TensorFlow Enterprise

Learn TensorFlow Enterprise

Release Date: 2020/11/01
ISBN: 9781800209145
Topic:
0%
18
Chapters
0-1
Hours read
0k
Total Words
    Start Reading Now    
       Add to Wishlist       
View table of contents

Use TensorFlow Enterprise with other GCP services to improve the speed and efficiency of machine learning pipelines for reliable and stable enterprise-level deployment

Key Features

  • Build scalable, seamless, and enterprise-ready cloud-based machine learning applications using TensorFlow Enterprise
  • Discover how to accelerate the machine learning development life cycle using enterprise-grade services
  • Manage Google's cloud services to scale and optimize AI models in production

Book Description

TensorFlow as a machine learning (ML) library has matured into a production-ready ecosystem. This beginner's book uses practical examples to enable you to build and deploy TensorFlow models using optimal settings that ensure long-term support without having to worry about library deprecation or being left behind when it comes to bug fixes or workarounds.

The book begins by showing you how to refine your TensorFlow project and set it up for enterprise-level deployment. You'll then learn how to choose a future-proof version of TensorFlow. As you advance, you'll find out how to build and deploy models in a robust and stable environment by following recommended practices made available in TensorFlow Enterprise. This book also teaches you how to manage your services better and enhance the performance and reliability of your artificial intelligence (AI) applications. You'll discover how to use various enterprise-ready services to accelerate your ML and AI workflows on Google Cloud Platform (GCP). Finally, you'll scale your ML models and handle heavy workloads across CPUs, GPUs, and Cloud TPUs.

By the end of this TensorFlow book, you'll have learned the patterns needed for TensorFlow Enterprise model development, data pipelines, training, and deployment.

What you will learn

  • Discover how to set up a GCP TensorFlow Enterprise cloud instance and environment
  • Handle and format raw data that can be consumed by the TensorFlow model training process
  • Develop ML models and leverage prebuilt models using the TensorFlow Enterprise API
  • Use distributed training strategies and implement hyperparameter tuning to scale and improve your model training experiments
  • Scale the training process by using GPU and TPU clusters
  • Adopt the latest model optimization techniques and deployment methodologies to improve model efficiency

Who this book is for

This book is for data scientists, machine learning developers or engineers, and cloud practitioners who want to learn and implement various services and features offered by TensorFlow Enterprise from scratch. Basic knowledge of the machine learning development process will be useful.

Table of Contents

  1. Learn TensorFlow Enterprise
  2. Why subscribe?
  3. Contributors
  4. About the author
  5. About the reviewers
  6. Packt is searching for authors like you
  7. Preface
    1. Who this book is for
    2. What this book covers
    3. To get the most out of this book
    4. Download the example code files
    5. Download the color images
    6. Conventions used
    7. Get in touch
  8. Section 1 – TensorFlow Enterprise Services and Features
  9. Chapter 1: Overview of TensorFlow Enterprise
    1. Understanding TensorFlow Enterprise
    2. TensorFlow Enterprise packages
    3. Configuring cloud environments for TensorFlow Enterprise
    4. Setting up a cloud environment
    5. Creating a Google Cloud Storage bucket
    6. Enabling APIs
    7. Creating a data warehouse
    8. Using TensorFlow Enterprise in AI Platform
    9. Accessing the data sources
    10. Cloud Storage Reader
    11. BigQuery Reader
    12. Persisting data in BigQuery
    13. Persisting data in a storage bucket
    14. Summary
  10. Chapter 2: Running TensorFlow Enterprise in Google AI Platform
    1. Setting up a notebook environment
    2. AI Platform Notebook
    3. Deep Learning Virtual Machine Image
    4. Deep Learning Container (DLC)
    5. Suggestions for selecting workspaces
    6. Easy parameterized data extraction from BigQuery
    7. Putting it together
    8. Summary
  11. Section 2 – Data Preprocessing and Modeling
  12. Chapter 3: Data Preparation and Manipulation Techniques
    1. Converting tabular data to a TensorFlow dataset
    2. Converting a BigQuery table to a TensorFlow dataset
    3. Converting distributed CSV files to a TensorFlow dataset
    4. Preparing an example CSV
    5. Building filename patterns with TensorFlow I/O
    6. Creating a dataset from CSV files
    7. Inspecting the dataset
    8. Handling image data for input pipelines
    9. Constructing a protobuf message
    10. Decoding TFRecord and reconstructing the image
    11. Handling image data at scale
    12. Executing the steps
    13. Reading TFRecord and displaying it as images
    14. Summary
  13. Chapter 4: Reusable Models and Scalable Data Pipelines
    1. Using TensorFlow Hub
    2. Applying models from TensorFlow Hub
    3. Creating a generator to feed image data at scale
    4. Reusing pretrained ResNet feature vectors
    5. Compiling the model
    6. Training the model
    7. Scoring with test images
    8. Leveraging the TensorFlow Keras API
    9. Data acquisition
    10. Solving a data science problem with the us_border_volumes table
    11. Selecting features and a target for model training
    12. Streaming training data
    13. Input to a model
    14. Model training
    15. Working with TensorFlow Estimators
    16. Data pipeline for TensorFlow Estimators
    17. Summary
  14. Section 3 – Scaling and Tuning ML Works
  15. Chapter 5: Training at Scale
    1. Using the Cloud TPU through AI Platform
    2. Installing the Cloud SDK
    3. Whitelisting access for reading training data and writing artifacts (alternative)
    4. Execution command and format
    5. Cloud command arguments
    6. Organizing the training script
    7. Data streaming pipeline
    8. Submitting the training script
    9. Working with models in TensorFlow Hub
    10. Using the Google Cloud GPU through AI Platform
    11. Summary
  16. Chapter 6: Hyperparameter Tuning
    1. Technical requirements
    2. Delineating hyperparameter types
    3. Understanding the syntax and use of Keras Tuner
    4. Using hp.Int for hyperparameter definition
    5. Using hp.Choice for hyperparameter definition
    6. Using hp.Float for hyperparameter definition
    7. Delineating hyperparameter search algorithms
    8. Hyperband
    9. Bayesian optimization
    10. Random search
    11. Submitting tuning jobs in a local environment
    12. Submitting tuning jobs in Google's AI Platform
    13. Summary
  17. Section 4 – Model Optimization and Deployment
  18. Chapter 7: Model Optimization
    1. Technical requirements
    2. Understanding the quantization concept
    3. Training a baseline model
    4. Preparing a full original model for scoring
    5. Preparing test data
    6. Loading test data
    7. Scoring a single image with a full model
    8. Scoring batch images with a full model
    9. Converting a full model to a reduced float16 model
    10. Preparing the reduced float16 model for scoring
    11. Scoring a single image with a quantized model
    12. Scoring a batch image with a quantized model
    13. Converting a full model to a reduced hybrid quantization model
    14. Preparing test data for scoring
    15. Mapping a prediction to a class name
    16. Scoring with a hybrid quantization model
    17. Scoring a single image
    18. Scoring batch images
    19. Converting a full model to an integer quantization model
    20. Training a full model
    21. Scoring with an integer quantization model
    22. Preparing a test dataset for scoring
    23. Scoring batch images
    24. Summary
  19. Chapter 8: Best Practices for Model Training and Performance
    1. Input handling for loading data
    2. Working with the generator
    3. TFRecord dataset – ingestion pipeline
    4. TFRecord dataset – feature engineering and training
    5. Regularization
    6. L1 and L2 regularization
    7. Adversarial regularization
    8. Summary
  20. Chapter 9: Serving a TensorFlow Model
    1. Technical requirements
    2. Running Local Serving
    3. Understanding TensorFlow Serving with Docker
    4. Downloading TensorFlow Serving Docker images
    5. Creating a new image with the model and serving it
    6. Scoring through the RESTful API
    7. Summary
  21. Other Books You May Enjoy
    1. Leave a review - let other readers know what you think
3.138.125.2