Image Home
Image People
Image Publication
Image Applications
Image Data & Code
Image Library
Image Seminar
Image Link
Image Album

Search LAMDA


Description: ISD (Instance Specific Distance) is a package for learning instance specific distance functions. By using this package, you are able to assign instance specific distances for each labeled examples as well as unlabeled examples. The key of ISD learning is metric propagation.


•D.-C. Zhan, M. Li, Y.-F. Li and Z.-H. Zhou. Learning instance specific distances using metric propagation. In: Proceedings of the 26th International Conference on Machine Learning (ICML'09), Montreal, Canada, 2009, pp.1225-1232.

ATTN: This package is free for academic usage. You can run it at your own risk. For other purposes, please contact Prof. Zhi-Hua Zhou (

Requirement: The package was developed with MATLAB (R2009a) and C-Mex files. The C/C++ routines can be called in MATLAB environment. In addition, mosek (v5.0) is used for solving the QP problem in ISD-L1.

Data format: Matlab data file which contains a matrix variable named "content" and each column is an instance. The last row/feature is the label index. frame1.m and frame2.m are entry functions for ISD-L1 and ISD-L2, respectively.

ATTN2: This package was developed by Mr. De-Chuan Zhan ( For any problem concerning the code, please feel free to contact Mr. Zhan.

Download: code (95Kb)
  Name Size

(for FireFox 3+ and IE 7+)
Contact LAMDA: (email) (tel) +86-25-89685926. © LAMDA, 2016