Skip to content

Caffe/Neon prototxt training file for our Neurocomputing2017 work: Fuzzy Quantitative Deep Compression Network

License

Notifications You must be signed in to change notification settings

cs-chan/Fuzzy-Compression

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

69 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Fuzzy Quantitative Deep Compression Network

Released on April, 2017

Description

This repository contains code and trained models of our Neurocomputing work with titled Fuzzy Quantitative Deep Compression Network.

Dependency

This repository requires CAFFE and/or Nervana Systems Neon to be installed.

  • To install Nervana System Neon, please visit: https://github.com/NervanaSystems/neon
  • To install CAFFE library, users have to contact the authors of the following paper to get the modified CAFFE:

Han, S., Pool, J., Tran, J., & Dally, W. (2015). Learning both weights and connections for efficient neural network. In Advances in Neural Information Processing Systems (pp. 1135-1143).

Citation

If you find this code useful for your research, please cite

@article{tan2017fuzzy,
  title={Fuzzy Qualitative Deep Compression Network},
  author={Tan, Wei Ren and Chan, Chee Seng and Aguirre, Hern{\'a}n E and Tanaka, Kiyoshi},
  journal={Neurocomputing},
  volume = {251},
  page = {1-15},
  year={2017},
  publisher={Elsevier}
}

Datasets

Codes for Wikiart dataset are written in CAFFE. This repository does not include the original Wikiart dataset used. Credit is given to the authors of the following paper for introducing the Wikiart dataset:

Saleh, B., & Elgammal, A. (2015). Large-scale Classification of Fine-Art Paintings: Learning The Right Metric on The Right Feature. arXiv preprint arXiv:1505.00855.

In order to replicate or to have a fair comparison to our work, users may access the "new" Wikiart dataset (It was splitted into training and validation sets) available at this https URL.

For the rest of the datasets, please visit:

For more details, please read the readme files in the subdirectories. We may release codes written in Tensorflow in the future.

Feedback

Suggestions and opinions of this work (both positive and negative) are greatly welcome. Please contact the authors by sending email to wrtan.edu at gmail.com or cs.chan at um.edu.my.

License

BSD-3, see LICENSE file for details.