Overview
This course aims to explore the work required to enable a TensorFlow deployment on open source Mesa drivers for GPU compute in AI/Machine Learning applications. The course covers topics such as TensorFlow deployment on GPUs, SYCL via LLVM and OpenCL, enabling OpenCL on a broader range of hardware, and the current state of the One API initiative. The teaching method involves discussing the theoretical background and practical applications of deploying TensorFlow on open source GPUs. This course is intended for individuals interested in AI/Machine Learning applications, GPU computing, and open source software development.
Syllabus
Introduction
Outline
Tensorflow
ICON
Building
Backend
Execution
OpenCL
Vulcan
Shader vs Kernel
One API initiative
Sickle
Level Zero
Current State
Malcolm in the Middle
Clover OpenCL
Conclusion
Taught by
linux.conf.au