Axen AI Accelerator

AXEN (ACCelerator ENabler) is my attempt to build an end-to-end AI stack.

The goal is to enable PyTorch AI application acceleration through custom PCIe AI Accelerator Hardware.

System Architecture

Installation

{to be written later}

PCIe Hardware Memory Mapping

There are two main regions:

• BAR0 (via MMIO)
• BAR1 (via DMA)

{to be written later}