Hands-On Machine Learning on Google Cloud Platform电子书

Title Page

Copyright and Credits

Hands-On Machine Learning on Google Cloud Platform

Packt Upsell

Why subscribe?

PacktPub.com

Contributors

About the authors

About the reviewers

Packt is searching for authors like you

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Reviews

Introducing the Google Cloud Platform

ML and the cloud

The nature of the cloud

Public cloud

Managed cloud versus unmanaged cloud

IaaS versus PaaS versus SaaS

Costs and pricing

ML

Introducing the GCP

Mapping the GCP

Getting started with GCP

Project-based organization

Creating your first project

Roles and permissions

Further reading

Summary

Google Compute Engine

Google Compute Engine

VMs, disks, images, and snapshots

Creating a VM

Google Shell

Google Cloud Platform SDK

Gcloud

Gcloud config

Accessing your instance with gcloud

Transferring files with gcloud

Managing the VM

IPs

Setting up a data science stack on the VM

BOX the ipython console

Troubleshooting

Adding GPUs to instances

Startup scripts and stop scripts

Resources and further reading

Summary

Google Cloud Storage

Google Cloud Storage

Box–storage versus drive

Accessing control lists

Access and management through the web console

gsutil

gsutil cheatsheet

Advanced gsutil

Signed URLs

Creating a bucket in Google Cloud Storage

Google Storage namespace

Naming a bucket

Naming an object

Creating a bucket

Google Cloud Storage console

Google Cloud Storage gsutil

Life cycle management

Google Cloud SQL

Databases supported

Google Cloud SQL performance and scalability

Google Cloud SQL security and architecture

Creating Google Cloud SQL instances

Summary

Querying Your Data with BigQuery

Approaching big data

Data structuring

Querying the database

SQL basics

Google BigQuery

BigQuery basics

Using a graphical web UI

Visualizing data with Google Data Studio

Creating reports in Data Studio

Summary

Transforming Your Data

How to clean and prepare the data

Google Cloud Dataprep

Exploring Dataprep console

Removing empty cells

Replacing incorrect values

Mismatched values

Finding outliers in the data

Visual functionality

Statistical information

Removing outliers

Run Job

Scale of features

Min–max normalization

z score standardization

Google Cloud Dataflow

Summary

Essential Machine Learning

Applications of machine learning

Financial services

Retail industry

Telecom industry

Supervised and unsupervised machine learning

Overview of machine learning techniques

Objective function in regression

Linear regression

Decision tree

Random forest

Gradient boosting

Neural network

Logistic regression

Objective function in classification

Data splitting

Measuring the accuracy of a model

Absolute error

Root mean square error

The difference between machine learning and deep learning

Applications of deep learning

Summary

Google Machine Learning APIs

Vision API

Enabling the API

Opening an instance

Creating an instance using Cloud Shell

Label detection

Text detection

Logo detection

Landmark detection

Cloud Translation API

Enabling the API

Natural Language API

Speech-to-text API

Video Intelligence API

Summary

Creating ML Applications with Firebase

Features of Firebase

Building a web application

Building a mobile application

Summary

Neural Networks with TensorFlow and Keras

Overview of a neural network

Setting up Google Cloud Datalab

Installing and importing the required packages

Working details of a simple neural network

Backpropagation

Implementing a simple neural network in Keras

Understanding the various loss functions

Softmax activation

Building a more complex network in Keras

Activation functions

Optimizers

Increasing the depth of network

Impact on change in batch size

Implementing neural networks in TensorFlow

Using premade estimators

Creating custom estimators

Summary

Evaluating Results with TensorBoard

Setting up TensorBoard

Overview of summary operations

Ways to debug the code

Setting up TensorBoard from TensorFlow

Summaries from custom estimator

Summary

Optimizing the Model through Hyperparameter Tuning

The intuition of hyperparameter tuning

Overview of hyperparameter tuning

Hyperparameter tuning in Google Cloud

The model file

Configuration file

Setup file

The __init__ file

Summary

Preventing Overfitting with Regularization

Intuition of over/under fitting

Reducing overfitting

Implementing L2 regularization

Implementing L1 regularization

Implementing dropout

Reducing underfitting

Summary

Beyond Feedforward Networks – CNN and RNN

Convolutional neural networks

Convolution layer

Rectified Linear Units

Pooling layers

Fully connected layer

Structure of a CNN

TensorFlow overview

Handwriting Recognition using CNN and TensorFlow

Run Python code on Google Cloud Shell

Recurrent neural network

Fully recurrent neural networks

Recursive neural networks

Hopfield recurrent neural networks

Elman neural networks

Long short-term memory networks

Handwriting Recognition using RNN and TensorFlow

LSTM on Google Cloud Shell

Summary

Time Series with LSTMs

Introducing time series

Classical approach to time series

Estimation of the trend component

Estimating the seasonality component

Time series models

Autoregressive models

Moving average models

Autoregressive moving average model

Autoregressive integrated moving average models

Removing seasonality from a time series

Analyzing a time series dataset

Identifying a trend in a time series

Time series decomposition

Additive method

Multiplicative method

LSTM for time series analysis

Overview of the time series dataset

Data scaling

Data splitting

Building the model

Making predictions

Summary

Reinforcement Learning

Reinforcement learning introduction

Agent-Environment interface

Markov Decision Process

Discounted cumulative reward

Exploration versus exploitation

Reinforcement learning techniques

Q-learning

Temporal difference learning

Dynamic Programming

Monte Carlo methods

Deep Q-Network

OpenAI Gym

Cart-Pole system

Learning phase

Testing phase

Summary

Generative Neural Networks

Unsupervised learning

Generative models

Restricted Boltzmann machine

Boltzmann machine architecture

Boltzmann machine disadvantages

Deep Boltzmann machines

Autoencoder

Variational autoencoder

Generative adversarial network

Adversarial autoencoder

Feature extraction using RBM

Breast cancer dataset

Data preparation

Model fitting

Autoencoder with Keras

Load data

Keras model overview

Sequential model

Keras functional API

Define model architecture

Magenta

The NSynth dataset

Summary

Chatbots

Chatbots fundamentals

Chatbot history

The imitation game

Eliza

Parry

Jabberwacky

Dr. Sbaitso

ALICE

SmarterChild

IBM Watson

Building a bot

Intents

Entities

Context

Chatbots

Essential requirements

The importance of the text

Word transposition

Checking a value against a pattern

Maintaining context

Chatbots architecture

Natural language processing

Natural language understanding

Google Cloud Dialogflow

Dialogflow overview

Basics Dialogflow elements

Agents

Intent

Entity

Action

Context

Building a chatbot with Dialogflow

Agent creation

Intent definition

Summary