Git Product home page Git Product logo

yodann's Introduction

YodaNN

Inspiration

The embedded systems evolved, and their purposes multiplied in the last couple of years. We expect our smartphones to run ML algorithms and the ASIC/FPGA entered fast the cloud processing industry. Image processing plays a role in automation and IoT. This feature can be used in medicine, security faster communication.

The inspiration was the article YodaNN: An Architecture for Ultralow Power Binary-Weight CNN Acceleration; researchers present a new architecture for embedded systems to support the performance needs in AI in the next couple of years.

What it does

Our project consists of a simulation of an FPGA board (Nexys A7) attached to an ARMv8 processor, similar to ones in smartphones or SBCs.The processor sends the pixels to the pre-decoder. This module determines in which SCM the data should be read/written. The operations execute in parallel with the main memory for optimization.

How I built it

As a platform, we used Vivado Webpack because it has many testing tools and a visual interface to create RTL designs. We developed the modules in Verilog. We followed the instructions and the diagrams from the article. link

Challenges I ran into

First of all, the debugging was quite challenging because of the size of the diagram. Also, the instructions were a little unclear, but we managed to create the non-shifting version.

Accomplishments that I'm proud of

We managed to simulate a simplified version of the image memory module from the new architecture. Also, we demonstrated that the system is faster than the standard SDRAM.

What I learned

This weekend we learned a lot of things about embedded systems, memory, computer architecture, etc. Also, we improved our Verilog and circuit design skills.

What's next for YodaNN

In the following period, we will improve the research on the subject and obtain an IP authorization. Also, we could collaborate with academic research teams / R&D centers from companies to build the module.

yodann's People

Contributors

cristinatomoescu avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.