Representing the World
The purpose of the Nx tensor abstraction is to give you a flexible data structure for representing the real world. Machine learning is applicable to nearly every field. To use machine learning for your applications, you need to choose an appropriate numerical representation for your data. Depending on the type of data you have, there’s likely already a common or natural way to represent the data as a tensor.
Tabular Data
Tabular data or structured data is data with a structure similar to what you would find in a relational database or CSV file. Each column represents a feature of the data, such as petal length in the iris example from Chapter 1, Make Machines That Learn. You’ll often see tabular data represented in a ...
Get Machine Learning in Elixir now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
