Overview Prerequisites Learning Outcomes Course Outline Related Courses

Overview

You’ll learn the fundamental tools and techniques for running GPU-accelerated Python applications in this workshop using CUDA and the NUMBA compiler GPUs. You’ll use a live, cloud-based GPU-enabled development environment to work though dozens of hands-on coding exercises.

Learn how to:

  • Write code for a GPU accelerator
  • Configure code parallelization using the CUDA thread hierarchy
  • Manage and optimize memory migration between the CPU and GPU accelerator
  • Generate random numbers on the GPU
  • Intermediate GPU memory management techniques

Finish by implementing your new workflow to accelerate a fully functional linear algebra program originally designed for CPUs to observe impressive performance gains.

After the workshop ends, you’ll have additional resources enabling you to create new GPU-accelerated applications on your own.

Prerequisites

  • Basic Python competency including familiarity with variable types, loops, conditional statements, functions, and array manipulations
  • NumPy competency including the use of ndarrays and ufuncs
  • No previous knowledge of CUDA programming is required

Learning Outcomes

Gain understanding on how to use fundamental tools and techniques for GPU-accelerate Python applications with CUDA and Numba, including:

  • GPU-accelerate NumPy ufuncs with a few lines of code
  • Write custom CUDA device kernels for maximum performance and flexibility
  • Use memory coalescing and on-device shared memory to increase CUDA kernel bandwidth

Why Deep Learning Institute Hands-on Training?

  • Learn how build deep learning and accelerated computing applications across a wide range of industry segments such as autonomous vehicles, digital content creation, finance, game development, healthcare, and more.
  • Learn in aguided, hands-on experience using the most widely-used, industry-standard software, tools, and frameworks.
  • Gain real world expertise through content designed in collaboration with industry leaders such as the Children’s Hospital of Los Angeles, Mayo Clinic, and PwC.
  • Earn NVIDIA DLI certification to prove your subject matter competency and support professional career growth.
  • Access content anywhere, anytime with a fully configured GPU-accelerated workstation in the cloud.

Certification

Upon successful completion of the workshop, participants will receive NVIDIA DLI Certification to recognize subject matter competency and support professional career growth.

Course Outline

Introduction

  • Getting started

The first 15 minutes introduces how to set-up your training environment.

Introduction to CUDA Python with Numba

  • Optimize CPU code with the Numba compiler
  • GPU-accelerate NumPy ufuncs
  • Optimize host-to-device and device-to-host memory transfers

Begin to working with the Numba compiler and CUDA programming in Python. Learn to use Numba decorators to accelerate numerical Python functions. Complete an assessment to accelerate a neural network layer.

Custom CUDA Kernels in Python with Numba

  • Learn CUDA’s parallel thread hierarchy
  • Launch massively parallel custom CUDA kernels on the GPU
  • Utilize atomic operations to avoid race conditions during parallel execution.

Learn how to extend parallel program possibilities,. including the ability to design and write flexible and powerful CUDA kernels. You’ll grasp ho to easily handle race conditions with CUDA atomic operations and parallel thread synchronization. You’ll also complete an assessment to accelerate a Mandelbrot set calculator and visualizer.

RNG, Multidimensional Grids, and Shared Memory for CUDA Python with Numba

  • Use xoroshiro128+ RNG to support GPU-accelerated monte carlo methods
  • Learn multidimensional grid creation and how to work in parallel on 2D matrices
  • Leverage on-device shared memory to communicate between threads

Generate a random number state for thousands of parallel threads in this intermediate-level module. Use shared memory for on-device caching and promoting memory coalescing while reshaping 2D matrices.

Assessment

  • Accelerate a CPU-only linear algebra subprogram

This module teaches you to leverage your learning to accelerate a CPU-only linear algebra subroutine for massive performance gains.

Next Steps

  • Complete workshop survey
  • Set up your own GPU enabled environment

Finally, learn how to set up your CUDA and GPU-enabled environment to begin work on your own projects.

  • Tools, Libraries, and Frameworks: Numba, NumPy

Related to this course

New NVFDLCV

Fundamentals of Deep Learning for Computer Vision

This workshop teaches deep learning techniques for a range of computer vision tasks through a series of hands-on exercises…

From £1,145 1 Day Classroom/Virtual
New NVFDLMDT

Fundamentals of Deep Learning for Multiple Data Types

This workshop uses a series of hands-on exercises to teach deep learning techniques for a range of problems involving mult…

From £1,145 1 Day Classroom/Virtual
New NVFDLNLP

Fundamentals of Deep Learning for Natural Language Processing

This workshop teaches deep learning techniques for understanding textual input using natural language processing (NLP) thr…

From £1,145 1 Day Classroom/Virtual
New NVDLAVP

Deep Learning for Autonomous Vehicles Perception

This workshop teaches you to apply deep learning techniques to design, train, and deploy deep neural networks for autonomo…

From £1,405 1 Day Classroom
New NVDLCCA

Deep Learning for Digital Content Creation with Autoencoders

This workshop teaches deep learning techniques to design, train, and deploy neural networks for digital content creation t…

From £1,405 1 Day Classroom
New NVDLCCG

Deep Learning for Digital Content Creation with GANs

This workshop teaches deep learning techniques to design, train, and deploy neural networks for digital content creation t…

From £1,405 1 Day Classroom
New NVDLHIA

Deep Learning for Healthcare Image Analysis

You’ll learn how to apply deep learning to radiology and medical imaging in this class. The workshop covers: …

From £1,405 1 Day Classroom
New NVDLIVA

Deep Learning for Intelligent Video Analytics

This hands-on workshop teaches you the foundational elements of intelligent video analytics (IVA) by integrating deep lear…

From £1,405 1 Day Classroom
New NVDLRTS

Deep Learning for Robotics

In this hand-on workshop, you’ll learn the fundamentals of integrating object detection into a mobile robot running …

From £1,405 1 Day Classroom