d.abstract: dataset and pre-processing abstraction

d.abstract is a -lightweight- library for defining and processing data(sets). The goals of the library is to:

1. uniformize dataset classes,

2. speed up experiments where the data does not fit into memory,

3. offer a general way of processing data independent of the data source,

4. seamlessly combine different datasets.

Readme

• Getting started
• Tutorials

Contents

• Data processing
• Data abstraction
• Dataset
• Utilities

Index

• Indices and tables