The Introduction to Graphics Processing course is designed to provide an overview of the graphics technology landscape: how it is changing, how and why each of these approaches is used/useful, and under what circumstances they make business/technology sense.
The course begins with participants learning to use FPGA/ASIC programming. Next, the course will cover environments/simulators and will also focus on deployment options, architectural choices, security issues, and other common problems. The course concludes with an introduction to the OpenCL/Vulkan ecosystem as a standards-based competitor to CUDA.
Purpose
| Learn a working knowledge of graphics processing practices within Machine Learning using CUDA, PyCuda, OpenCL, Vulkan, and Tensorflow |
Audience
| Developers who need to skill up quickly on Machine Learning and Graphics Processing technologies |
Role
| Software Developer |
Skill Level
| Introduction |
Style
| Lecture | Hands-on Activities | Labs |
Duration
| 3 Days |
Related Technologies
| Machine Learning | CUDA | Python | Tensorflow |
Â
Learning Objectives
- Describe the technology landscape including use of CPUs, GPUs, FPGA, ASIC, and CUDA
- Use Python Bindings to build, run and debug
- Compare OpenCL and Vulkan and describe reasons for using
- Identify how TensorFlow can accelerate machine-learning activities
- Discuss cloud-based options for accelerating GPU
