Hugging Face's logo

Ruichen0424
/
AB-BNN

English
Model card Files
xet
Community
Ruichen0424 commited on
Commit
276cabc
·
verified ·
1 Parent(s): 196ee87

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -19,6 +19,7 @@ base_model:
19
  [![Paper](https://img.shields.io/badge/Arxiv-2403.03739-B31B1B.svg?style=flat-square)](https://arxiv.org/abs/2403.03739)
20
  [![CVPR 2024](https://img.shields.io/badge/CVPR%202024-Poster-4b44ce.svg?style=flat-square)](https://cvpr.thecvf.com/virtual/2024/poster/29447)
21
  [![Google Scholar](https://img.shields.io/badge/Google%20Scholar-Paper-4285F4?style=flat-square&logo=google-scholar&logoColor=white)](https://scholar.google.com/scholar?cluster=9219398500921383941)
 
22
  [![GitHub](https://img.shields.io/badge/GitHub-Repository-black?logo=github)](https://github.com/Ruichen0424/AB-BNN)
23
 
24
  </div>
@@ -29,6 +30,8 @@ This repository contains the **pre-trained weights** for the paper **"A&B BNN: A
29
 
30
  **A&B BNN** proposes to directly remove part of the multiplication operations in a traditional BNN and replace the rest with an equal number of bit operations. It introduces the mask layer and the quantized RPReLU structure based on the normalizer-free network architecture.
31
 
 
 
32
  ### ✨ Key Highlights
33
  * **Hardware-Friendly**: Removes multiplication operations, replacing them with bit operations.
34
  * **Competitive Performance**: Achieves **92.30%**, **69.35%**, and **66.89%** on CIFAR-10, CIFAR-100, and ImageNet respectively.
 
19
  [![Paper](https://img.shields.io/badge/Arxiv-2403.03739-B31B1B.svg?style=flat-square)](https://arxiv.org/abs/2403.03739)
20
  [![CVPR 2024](https://img.shields.io/badge/CVPR%202024-Poster-4b44ce.svg?style=flat-square)](https://cvpr.thecvf.com/virtual/2024/poster/29447)
21
  [![Google Scholar](https://img.shields.io/badge/Google%20Scholar-Paper-4285F4?style=flat-square&logo=google-scholar&logoColor=white)](https://scholar.google.com/scholar?cluster=9219398500921383941)
22
+ [![IEEE](https://img.shields.io/badge/IEEE-Paper-00629B?style=flat-square&logo=ieee&logoColor=white)](https://xploreqa.ieee.org/document/10656026)
23
  [![GitHub](https://img.shields.io/badge/GitHub-Repository-black?logo=github)](https://github.com/Ruichen0424/AB-BNN)
24
 
25
  </div>
 
30
 
31
  **A&B BNN** proposes to directly remove part of the multiplication operations in a traditional BNN and replace the rest with an equal number of bit operations. It introduces the mask layer and the quantized RPReLU structure based on the normalizer-free network architecture.
32
 
33
+ ![Poster](./assets/poster.png)
34
+
35
  ### ✨ Key Highlights
36
  * **Hardware-Friendly**: Removes multiplication operations, replacing them with bit operations.
37
  * **Competitive Performance**: Achieves **92.30%**, **69.35%**, and **66.89%** on CIFAR-10, CIFAR-100, and ImageNet respectively.