The Data Parallel Essentials for Python learning path series demonstrates high-performing code targeting Intel® XPUs using Python. The series will introduce Numba-dppy and show examples of how to write data-parallel code inside @numba.jit-decorated and @kernel decorator functions to offload them to a SYCL device. The series also covers dpctl, a companion library intended to make it easier to write Python native extensions based on SYCL. Numba-dppy is packaged as part of Intel®Distribution for Python, which is included with the Intel®oneAPI AI Analytics Toolkit. There are three modules in this series.
The virtual series is scheduled Wednesdays from 1:30 - 3:30 p.m. US Central.
- Module 1: Apr. 13, 2022
- Module 2: May 11, 2022
- Module 3: June 8, 2022
Module 1 (4/13): Introduction to Data Parallel Essentials for Python. This module will introduce basics of Numba; Numba-dppy with examples of how to write data-parallel code inside numba.jit-decorated functions and offload them to a SYCL device; writing an explicit kernel using the @numba_dppy.kernel decorator; and dpctl, a companion library for writing Python native extensions based on SYCL.
Module 2 (5/11): Introduction to Pairwise Distance Algorithm and Black-Scholes using Data Parallel Essentials for Python. In this module we will demonstrate [email protected] JIT method and using the @kernel decorator in two example applications that solve the Pairwise Distance and Black-Scholes problems.
Module 3 (6/8): Introduction to K-Means algorithm and GPairs algorithm using Data Parallel Essentials for Python. In this module we will demonstrate the @Numba JIT method and using the @kernel decorator in two additional example applications that implement the K-Means and GPairs algorithms.
Please see the Event Website linked below for a detailed Agenda of each Module.
Intel’s DevCloud will be used during the Learning Path Series. If you do not already have a DevCloud account, please visit this link to sign up prior to the first session.