Triton Deep Learning Kernel

This repository contains some useful kernels written in the Triton language for acceleration.

File Struture

+--notebook
| +--Jupyter Notebooks for Kernel tests
+--trace
| +-Torch trace files for performance benchmark
+--kernel
| +--Contains kernels and utils

Supported Kernels

Click to view the documentation of each kernel

• General Matrix Multiplication
• DropConnect