Go Machine Learning Projects : Eight Projects Demonstrating End-To-end Machine Learning and Predictive Analytics Applications in Go.

Print version record.

Cover; Title Page; Copyright and Credits; About Packt; Contributors; Table of Contents; Preface; Chapter 1: How to Solve All Machine Learning Problems; What is a problem? ; What is an algorithm? ; What is machine learning? ; Do you need machine learning?; The general problem solving process; What is a model?; What is a good model?; On writing and chapter organization ; Why Go? ; Quick start; Functions; Variables; Values ; Types ; Methods ; Interfaces; Packages and imports; Let's Go! ; Chapter 2: Linear Regression -- House Price Prediction; The project; Exploratory data analysis

Ingestion and indexingJanitorial work; Encoding categorical data; Handling bad numbers; Final requirement; Writing the code; Further exploratory work; The conditional expectation functions; Skews; Multicollinearity; Standardization; Linear regression; The regression; Cross-validation; Running the regression; Discussion and further work; Summary; Chapter 3: Classification -- Spam Email Detection; The project ; Exploratory data analysis ; Tokenization; Normalizing and lemmatizing; Stopwords; Ingesting the data; Handling errors; The classifier; Naive Bayes; TF-IDF ; Conditional probability

FeaturesBayes' theorem; Implementating the classifier; Class; Alternative class design; Classifier part II; Putting it all together; Summary; Chapter 4: Decomposing CO2 Trends Using Time Series Analysis; Exploratory data analysis; Downloading from non-HTTP sources; Handling non-standard data; Dealing with decimal dates; Plotting; Styling; Decomposition; STL; LOESS; The algorithm; Using STL; How to lie with statistics; More plotting; A primer on Gonum plots; The residuals plotter; Combining plots; Forecasting; Holt-Winters; Summary; References

Chapter 5: Clean Up Your Personal Twitter Timeline by Clustering TweetsThe project ; K-means ; DBSCAN; Data acquisition; Exploratory data analysis; Data massage; The processor ; Preprocessing a single word ; Normalizing a string; Preprocessing stopwords; Preprocessing Twitter entities ; Processing a single tweet ; Clustering ; Clustering with K-means ; Clustering with DBSCAN ; Clustering with DMMClust ; Real data; The program ; Tweaking the program; Tweaking distances ; Tweaking the preprocessing step ; Summary; Chapter 6: Neural Networks -- MNIST Handwriting Recognition; A neural network

Emulating a neural networkLinear algebra 101; Exploring activation functions; Learning; The project; Gorgonia; Getting the data; Acceptable format; From images to a matrix; What is a tensor?; From labels to one-hot vectors; Visualization; Preprocessing; Building a neural network; Feed forward; Handling errors with maybe; Explaining the feed forward function; Costs; Backpropagation; Training the neural network; Cross-validation; Summary; Chapter 7: Convolutional Neural Networks -- MNIST Handwriting Recognition; Everything you know about neurons is wrong ; Neural networks -- a redux; Gorgonia

Why?

Go is a highly preferred language for machine learning. The code is close to how it's actually executed in the machine. Over the course of this book, you will learn how to express complex ideas found in machine learning literature and implement them. You will also learn how to structure problems to solve them using machine learning with Go.

Includes bibliographical references.

eBooks on EBSCOhost EBSCO eBook Subscription Academic Collection - Worldwide