资源分类
》 资源 》算法》BinaryNet-Caffe2

BinaryNet-Caffe2

2020-02-04 | |  36 |   0 |   0

BNN_Caffe2

Quantized Neural Network on Caffe2

Goal: Porting QNN(BNN) from Theano version to Caffe2 version.

TODO:

  •  MNIST with LeNet

  •  MNIST with layers from QNN paper, without Batch Normalization, without Binarization.

  •  MNIST with layers from QNN paper, with Batch Normalization, without Binarization.

  •  Binarization functions (in progress)

  •  Layers with Binarization

  •  MNIST with layers from QNN paper, with Batch Normalization, with Binarization.

  •  Testing and debugging

  •  Training / evaluation on other dataset

Acknowledgment

Refs.

QNN Theano version paper https://arxiv.org/abs/1609.07061

Theano version github https://github.com/MatthieuCourbariaux/BinaryNet

License


上一篇：Accurate-Binary-Convolution-Network

下一篇：NetBinaryFormatterParser

用户评价
全部评价

热门资源

  • seetafaceJNI

    项目介绍 基于中科院seetaface2进行封装的JAVA...

  • spark-corenlp

    This package wraps Stanford CoreNLP annotators ...

  • Keras-ResNeXt

    Keras ResNeXt Implementation of ResNeXt models...

  • capsnet-with-caps...

    CapsNet with capsule-wise convolution Project ...

  • inferno-boilerplate

    This is a very basic boilerplate example for pe...