User Tools

Site Tools

You are here: start » rvkde
Trace:
rvkde

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision		Previous revision
rvkde [2008/08/06 15:02] – created dirtyrvkde [2008/08/19 00:55] (current) – dirty
Line 1: Line 1:
-====== Abstract ====== +====== About ====== 
-RVKDE, standing for Relaxed Variable Kernel Density Estimation, is an +Here is the start page of the wiki of [[http://mbi.ee.ncku.edu.tw/rvkde/|RVKDE]].  **RVKDE**, standing for Relaxed Variable Kernel Density Estimation, is an integrated software of classification, regression and density estimation.  Please read the COPYRIGHT file before using RVKDE.
-integrated software of classification, regression and density estimation. This +
-document explains the use of RVKDE.+
  
-RVKDE is available at 'http://mbi.ee.ncku.edu.tw/rvkde/'. Please read the +If you want to join us, e-mail to [[darby@ee.ncku.edu.tw|Tien-Hao Chang]] for a privileged account and you can start to edit this wiki.
-COPYRIGHT file before using RVKDE.+
  
-====== Table of contents ====== +====== Download ====== 
-  - Quick start +All releases of RVKDE are available at [[rvkde:download|this page]].
-  - Installation and data format +
-  - //kde-train.pl// usage +
-  - //kde-predict.pl// usage +
-  - Additional tools +
-  - Additional information +
- +
-====== Quick start ====== +
-If you are new to SVM and if the data is not large, please go to //tool// +
-directory and use //kde-easy.pl// after installation. It does everything, from +
-data scaling to parameter selection, automatic. +
- +
-Usage: kde-easy.pl TRAINING_FILE [TESTING_FILE] +
- +
-More information about parameter selection can be found in //tool/README//. +
- +
-====== Installation and data format ====== +
-**On Linux systems** +
-  - Donwload the latest version from the official site of RVKDE.<code> +
-wget http://mbi.ee.ncku.edu.tw/rvkde/res/rvkde-current-linux32.tgz +
-</code> +
-  - Extract the archive.<code> +
-tar zxvf rvkde-current-linux32.tgz +
-</code> +
-  - Execute them without arguments to show the usages.<code> +
-kde-train.pl +
-kde-predict.pl +
-</code> +
- +
--------------------------------------------------------------------------------- +
-**On Windows systems** +
-Before installing RVKDE, please make sure that Perl is installed. +
-[[http://www.activestate.com/|ActivePerl]] is a good Perl Distribution for +
-Windows. +
-  - Donwload the latest version from the official site of RVKDE.<code> +
-wget http://mbi.ee.ncku.edu.tw/rvkde/res/rvkde-current-win32.tgz +
-</code> +
-  - Extract the archive using a compressor (e.g. 7-zip, WinRAR or WinZip).  +
-  - Invoke the Windows console.<code> +
-[Start] -> [Run] -> Type 'cmd' -> [OK]<code> +
-</code> +
-  - Execute them without arguments to show the usages. +
-  - Using //cd// to the directory of your extracted RVKDE. +
-  - Execute them without arguments to show the usages.<code> +
-kde-train.pl +
-kde-predict.pl +
-</code> +
- +
-The format of training and testing data file is: +
- +
-<file> +
-<value> <index1>:<feature1> <index2>:<feature2> ... +
-. +
-. +
-. +
-</file> +
- +
-Each line contains an instance and is ended by a '\n' character.  For +
-classification, <value> is an integer indicating the class label (multi-class +
-is supported). For regression, <value> is the target value which can be any +
-real number. <index>:<feature> gives a feature (attribute) value. <index> is an +
-integer starting from 1 and <feature> is a real number. Labels in the testing +
-file are only used to calculate accuracy or errors. If they are unknown, just +
-fill the first column with any numbers. +
- +
-A sample classification data included in this package is **heart.scale**. +
-Type //kde-train.pl heart.scale//, and the program will read the training +
-data and output the model file //heart.scale.model//. If you have a test set +
-called //heart.scale.t//, then type //kde-predict.pl heart.scale.t +
-heart.scale.model output// to see the prediction accuracy. The //output// file +
-contains the predicted class labels.<code> +
-kde-train.pl heart.scale # train +
-kde-predict.pl heart.scale.t heart.scale.model output # predict +
-vi output # predicted results +
-</code> +
- +
-====== kde-train.pl usage ====== +
-<code> +
-Usage: kde-train.pl [OPTION]... TRAINING_FILE [MODEL_FILE] +
-kde-train.pl is a Perl wrapper for RVKDE to make the using of RVKDE like +
-LIBSVM. Make sure the rvkde binary executable file is in the same directory +
-of this script. +
- +
-Options: +
-  -a  alpha: set alpha in kernel function (default: 1) +
-  -b  beta: set beta in kernel function (default: 1) +
-  -h  print this help, then exit +
-  -s  ks: set maximum ks in kernel function (default: 10) +
-  -t  kt: set kt in kernel function (default: 10) +
-  -v  n: n-fold cross validation mode +
- +
-Examples: +
-  kde-train.pl heart.scale # generate heart.scale.model +
-  kde-train.pl heart.scale heart.scale.kde-model # different model name +
-  kde-train.pl -v 5 heart.scale # five-fold cross validation +
-  kde-train.pl -s 30 heart.scale # using different parameters +
- +
-Option -v splits the data into n parts by modulo arithmetic and calculates +
-cross validation performance on them. +
- +
-Report bugs to <darby@ee.ncku.edu.tw>. +
-</code> +
- +
-====== kde-predict.pl usage ====== +
-<code> +
-Usage: kde-predict.pl [OPTION]... TESTING_FILE MODEL_FILE OUTPUT_FILE +
-kde-predict.pl is a Perl wrapper for RVKDE to make the using of RVKDE like +
-LIBSVM. Make sure the rvkde binary executable file is in the same directory +
-of this script. +
- +
-Options: +
-  -a  alpha: set alpha in kernel function (default: 1) +
-  -b  beta: set beta in kernel function (default: 1) +
-  -h  print this help, then exit +
-  -s  ks: set ks in kernel function (default: 10) +
-  -t  kt: set kt in kernel function (default: 10) +
- +
-  TESTING_FILE is the testing data you want to predict. +
-  MODEL_FILE is the model file generated by kde-train.pl. +
-  OUTPUT_FILE contains the predicted results by RVKDE. +
- +
-Examples: +
-  kde-predict.pl heart.scale.t heart.scale.model output +
-  kde-predict.pl -s 30 -t 30 heart.scale.t heart.scale.model output +
- +
-Unlink LIBSVM, RVKDE is a lazy-learning tool. The model file contains only +
-basic information and the real prediction model is constructed right before the +
-prediction operation. It means you can specify different parameters (alpha, +
-beta, ks and kt) for kde-predict.pl as those specified for kde-train.pl. Only +
-one condition must be hold: the ks for kde-predict.pl is less than or equal to +
-the ks for kde-train.pl. +
- +
-Report bugs to <darby@ee.ncku.edu.tw>. +
-</code> +
- +
-====== Additional tools ====== +
-See the README file in the tool directory. +
- +
-====== Additional information ====== +
-If you find RVKDE helpful, please cite it as +
- +
-Yen-Jen Oyang, Shien-Ching Hwang, Yu-Yen Ou, Chien-Yu Chen, and Zhi-Wei Chen. +
-Data Classification with the Radial Basis Function Network Based on a Novel +
-Kernel Density Estimation Algorithm. //IEEE Transactions on Neural Networks//, +
-**16**, 225-236, 2005. +
- +
-Software available at 'http://mbi.ee.ncku.edu.tw/rvkde/'. +
- +
-For any questions and comments, please email darby@ee.ncku.edu.tw +
- +
-Acknowledgments: +
-This work was supported in part by the National Science Council of Taiwan via +
-the grant NSC 92-2213-E-002-095. The authors thank their group members and +
-users for many helpful discussions and comments.+
  
 +====== How to start? ======
 +  - [[rvkde:readme|README]]\\ You can start to use RVKDE by following this README. It includes a [[rvkde:readme#quick_start|quick start]] section and a [[rvkde:readme#installation_and_data format|installation]] section.
 +  - [[rvkde:usage|Usage]]\\ The usage section in the README only covers two wrapper scripts (__kde-train.pl__ and __kde-predict.pl__) for users that are familiar with [[http://www.csie.ntu.edu.tw/~cjlin/libsvm/|LIBSVM]].  This document demonstrates how to use the RVKDE binary executable file (__rvkde__ or __rvkde.exe__ on Windows system) directly.
rvkde.1218034926.txt.gz · Last modified: by dirty

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial-Share Alike 4.0 International
CC Attribution-Noncommercial-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki

Copyright © 2006-2014 MBI Lab. All rights reserved.

  • Service
  • DBD2BS: connecting a DNA-binding protein with its binding sites
  • E1DS: predicting catalytic sites based on 1-dimensional signatures
  • PRASA: an integrated web server that analyzes protein interaction types
  • Prote2s: predicting protein secondary structures
  • seeMotif: exploring and visualizing sequence motifs in 3D structures
  • MEDock: a docking web server for efficient prediction of protein-ligand binding sites
  • Protemot: predicting protein binding sites with automatically extracted geometrical templates
  • YGA: identifying distinct biological features between yeast gene sets
  • Database
  • AH-DB: collecting protein structure pairs before and after bindingan hot
  • YPA: collecting promoter features in Saccharomyces cerevisiae
  • Collaborative
  • iPDA: a integrated protein disorder analyzer
  • ProteDNA: Predictor of sequence-specific DNA-binding residues in transcription factors
  • MAGIIC-PRO: discovering functional long motifs that are interleaved by several large irregular gaps