Machine Learning with Go Quick Start Guide电子书

About Packt

Why subscribe?

Packt.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 Machine Learning with Go

What is ML?

Types of ML algorithms

Supervised learning problems

Unsupervised learning problems

Why write ML applications in Go?

The advantages of Go

Go's mature ecosystem

Transfer knowledge and models created in other languages

ML development life cycle

Defining problem and objectives

Acquiring and exploring data

Selecting the algorithm

Preparing data

Training

Validating/testing

Integrating and deploying

Re-validating

Summary

Further readings

Setting Up the Development Environment

Installing Go

Linux, macOS, and FreeBSD

Windows

Running Go interactively with gophernotes

Example – the most common phrases in positive and negative reviews

Initializing the example directory and downloading the dataset

Loading the dataset files

Parsing contents into a Struct

Loading the data into a Gota dataframe

Finding the most common phrases

Example – exploring body mass index data with gonum/plot

Installing gonum and gonum/plot

Loading the data

Understanding the distributions of the data series

Example – preprocessing data with Gota

Loading the data into Gota

Removing and renaming columns

Converting a column into a different type

Filtering out unwanted data

Normalizing the Height, Weight, and Age columns

Sampling to obtain training/validation subsets

Encoding data with categorical variables

Summary

Further readings

Supervised Learning

Classification

A simple model – the logistic classifier

Measuring performance

Precision and recall

ROC curves

Multi-class models

A non-linear model – the support vector machine

Overfitting and underfitting

Deep learning

Neural networks

A simple deep learning model architecture

Neural network training

Regression

Linear regression

Random forest regression

Other regression models

Summary

Further readings

Unsupervised Learning

Clustering

Principal component analysis

Summary

Further readings

Using Pretrained Models

How to restore a saved GoML model

Deciding when to adopt a polyglot approach

Example – invoking a Python model using os/exec

Example – invoking a Python model using HTTP

Example – deep learning using the TensorFlow API for Go

Installing TensorFlow

Import the pretrained TensorFlow model

Creating inputs to the TensorFlow model

Summary

Further readings

Deploying Machine Learning Applications

The continuous delivery feedback loop

Developing

Testing

Deployment

Dependencies

Model persistence

Monitoring

Structured logging

Capturing metrics

Feedback

Deployment models for ML applications

Infrastructure-as-a-service

Amazon Web Services

Microsoft Azure

Google Cloud

Platform-as-a-Service

Amazon Web Services

Amazon Sagemaker

Amazon AI Services

Microsoft Azure

Azure ML Studio

Azure Cognitive Services

Google Cloud

AI Platform

AI Building Blocks

Summary

Further readings

Conclusion - Successful ML Projects

When to use ML

Typical stages in a ML project

Business and data understanding

Data preparation

Modelling and evaluation

Deployment

When to combine ML with traditional code

Summary

Further readings

Other Books You May Enjoy

Leave a review - let other readers know what you think