Skip to content

Latest commit

 

History

History
26 lines (13 loc) · 1.04 KB

File metadata and controls

26 lines (13 loc) · 1.04 KB

Binarized Convolution Neural Network

TensorFlow 2.2 Python 3.6

Dataset: Kannada-MNIST

This project contains Tensorflow-Keras implementations of the BinaryNet and XNORNet papers:

Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1

XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks

This repo is fork of BinaryNet_and_XNORNet with modifications to support newer version TF 2.x with TF-Lite.

Note: TF-Lite support for these model is still in progress.

Result

XNOR Net on test dataset

Contribution is appreciated