readme

基于稀疏网络的精选机器学习模型

SNoW - Sparse Network of Winnows
Version 3.2.0
For questions or bug reports about the SNoW software, contact Nick Rizzolo
([e-mail])

University of Illinois at Urbana-Champaign
Department of Computer Science
Dr. Dan Roth - Cognitive Computation Group
[e-mail]

The makefile included with this distribution most probably requires GNU make.
In general, just type 'gmake' to compile the system, and 'gmake clean' to
remove all object files and binaries.  If you want to include SNoW's server
functionality in your build, use 'gmake SERVER=1'.  But do take a look at the
Makefile - you may find useful options or platform specific compilation
instructions.  (Just commenting or uncommenting a line or two.)  

The SNoW User's Guide is no longer included in the software distribution.  You
can download it separately from the webpage.  As of 8/1/04, the User's Guide
is completely current with respect to version  3.1.4.  The revision history
can also be a useful source of information about SNoW's new features.


Packing list:

0README                         : This file
revisions.txt                   : SNoW's revision history

booleanexample/booleantest.snow       : See SNoW in action, learning a simple
booleanexample/one_target.arch        :  boolean concept with these files.  Run
booleanexample/one_target_train.snow  :  "run.bat" if your platform is WIN32, or
booleanexample/run.bat                :  "run.sh" on UNIX.
booleanexample/run.sh*                :
booleanexample/two_targets.arch       :
booleanexample/two_targets_train.snow :

booleanexample/conjunction_generator.pl* : Scripts for enumerating all the
booleanexample/disjunction_generator.pl* :  examples in simple boolean concepts.

tutorial/example_client.pl*     : An example client for SNoW's server mode
tutorial/archfile               : More useful example files.
tutorial/testdata.feat          :
tutorial/traindata.feat         :
tutorial/tutorial.bat           :
tutorial/tutorial.sh*           :
tutorial/Getopt/Declare.pm      :

Makefile                        : Used to compile the software

Cloud.h
Example.h
GlobalParams.h
LearningAlgorithm.h
MinimalFeature.h
NaiveBayes.h
Network.h
Perceptron.h
SendReceive.h
Snow.h
SnowParam.h
SnowTypes.h
Target.h
TargetIdSet.h
TargetRank.h
Usage.h
Winnow.h

Cloud.cpp
Example.cpp
Main.cpp
NaiveBayes.cpp
Network.cpp
Perceptron.cpp
Snow.cpp
SnowParam.cpp
Target.cpp
TargetIdSet.cpp
TargetRank.cpp
Winnow.cpp
getopt.c