oneDAL

This document specifies requirements for implementations of oneAPI Data Analytics Library (oneDAL).

oneDAL is a library that helps speed up big data analysis by providing highly optimized algorithmic building blocks for all stages of data analytics (preprocessing, transformation, analysis, modeling, validation, and decision making) in batch, online, and distributed processing modes of computation. The current version of oneDAL provides Data Parallel C++ (DPC++) API extensions to the traditional C++ interface.

For general information, visit oneDAL GitHub* page.

• Introduction
• Glossary
• Mathematical Notations
• Programming model
  • End-to-end example
  • Descriptors
  • Operations
  • Computational modes
• Common Interface
  • Current Version of this oneDAL Specification
  • Header files
  • Namespaces
  • Error handling
  • Common type definitions
• Data management
  • Array
  • Accessors
    • Column accessor
    • Row accessor
  • Data Sources
    • CSV data source
  • Tables
    • Homogeneous table
• Algorithms
  • Clustering
    • K-Means
    • K-Means initialization
  • Nearest Neighbors (kNN)
    • k-Nearest Neighbors Classification (k-NN)
  • Decomposition
    • Principal Components Analysis (PCA)
• Appendix
  • k-d Tree
• Bibliography