Reference Number: DEVCOM-134

Project Description

Students will develop user-interfaces for the Neural Place and Route (NPR) tool that is used to map an ONNX model of a deep neural network into a Field Programmable Neural Array (FPNA) chip architecture. This will include 1) interface for selecting neural network and IO files, 2) visualization of the neural network in the tool, 3) visualization of the FPNA chip architecture, 4) visualization of the connectivity of compute elements (CEs) within the FPNA chip once the chip is mapped.

Technical Skills

• Strong debugging skills.
• Proficiency in Python coding, particularly in packages w.r.t. user interfaces and graphics (Pillow, Qt, Tkinter, etc.).
• Understanding of deep neural networks (DNNs) preferred, helps to understand the goals of NPR.