Data Parallel Control (dpctl) 0.20.0dev0 documentation (original) (raw)

Back to top

View this page

Toggle table of contents sidebar

Python package dpctl enables Python users to engage multiple compute devices commonly available in modern consumer- and server-grade computers using industry-standard SYCL execution modelfacilitated by SYCL 2020 standard-compliant Intel(R) oneAPI DPC++ compiler.

dpctl provides a reference data-parallel implementation of array library dpctl.tensor conforming to Python Array API specification. The implementation adheres to a programming model affording clear control over the compute device where array computations and memory allocations take place.

Beginner Guides

New to dpctl? Check out the Tutorials. They are a hands-on introduction for beginners.

User Guides

The user guides are recipes for key tasks and common problems.

API Reference

API Reference contains detailed documentation of functionality provided in dpctl and its components.

Contibutor Guides

The contributing guidelines will suggest a process of contributing to dpctl.