计算机科学论坛--matlab的数据挖掘工具箱spider.mht
来自「matlab datming spider toolbox」· MHT 代码 · 共 1,373 行 · 第 1/5 页
MHT
1,373 行
border=3D0></A> <A=20
=
href=3D"http://www.ieee.org.cn/reannounce.asp?BoardID=3D69&replyID=3D=
105586&id=3D51528&star=3D1&reply=3Dtrue"><IMG=20
alt=3D=D2=FD=D3=C3=BB=D8=B8=B4=D5=E2=B8=F6=CC=F9=D7=D3 =
src=3D"http://www.ieee.org.cn/pic/reply.gif"=20
border=3D0></A> <A=20
=
href=3D"http://www.ieee.org.cn/reannounce.asp?BoardID=3D69&replyID=3D=
105586&id=3D51528&star=3D1"><IMG=20
alt=3D=BB=D8=B8=B4=D5=E2=B8=F6=CC=F9=D7=D3 =
src=3D"http://www.ieee.org.cn/pic/reply_a.gif"=20
border=3D0></A> <A=20
=
href=3D"http://blogger.org.cn/blog/blog.asp?name=3DDMman"><IMG=20
alt=3D=B2=E9=BF=B4DMman=B5=C4=B2=A9=BF=CD =
src=3D"http://www.ieee.org.cn/images/boke.gif"=20
border=3D0></A></TD>
<TD width=3D50><B>=C2=A5=D6=F7</B></TD></TR>
<TR>
<TD bgColor=3D#6595d6 colSpan=3D2 =
height=3D1></TD></TR></TBODY></TABLE>
<DIV id=3Dgoogleadlink1 height=3D"87" width=3D"490"></DIV>
<BLOCKQUOTE>
<TABLE class=3Dtablebody2=20
style=3D"TABLE-LAYOUT: fixed; WORD-BREAK: break-all" =
width=3D"90%"=20
border=3D0><TBODY>
<TR>
<TD style=3D"FONT-SIZE: 9pt; LINE-HEIGHT: 12pt" =
width=3D"100%"><IMG=20
alt=3D=B7=A2=CC=F9=D0=C4=C7=E9 =
src=3D"http://www.ieee.org.cn/face/face1.gif"=20
=
border=3D0> <B>matlab=B5=C4=CA=FD=BE=DD=CD=DA=BE=F2=B9=A4=BE=DF=CF=E4=
spider</B><BR>
<DIV width=3D"100%">
<DIV style=3D"FLOAT: right"></DIV>
<DIV>=D2=BB spider=D6=F7=D2=B3<A class=3Dcontentlink=20
href=3D"http://www.kyb.mpg.de/bs/people/spider/"=20
=
target=3D_blank>http://www.kyb.mpg.de/bs/people/spider/</A> =A3=A8=D2=
=B2=BF=C9=D2=D4=D4=DAgoogle=C9=CF=CB=D1=CB=F7spider matlab=B5=C3=B5=BD=
=A3=A9=A3=AC=B9=D8=D3=DA=CB=FC=B5=C4=BD=E9=C9=DC=BF=C9=D2=D4=B2=CE=BF=BC=CD=
=F8=D6=B7=D7=CA=C1=CF
<P></P>
=
<P>=B6=FE =CA=B9=D3=C3=CA=B1=CE=AAmatlab+spider+Weka=A3=BB=D2=F2=CE=AA=
spider=D6=D0=B5=C4=D2=BB=D0=A9=CB=E3=B7=A8=D2=FD=D3=C3=C1=CBWeka=A3=AC=B1=
=C8=C8=E7j48</P>
<P>=B0=B2=D7=B0=D7=A2=D2=E2=A3=BA</P>
<P>1 matlab7=A3=A8R14=A3=A9</P>
=
<P> 6.5=B0=E6=B1=BE=B6=D4java=B5=C4=D6=A7=B3=D6=B2=BB=B9=BB=A3=
=AC=BB=B9=C3=BB=D3=D0=BF=AA=B7=A2javaclasspath=B5=C8=BA=AF=CA=FD</P>
=
<P>??? Undefined function or variable 'javaclass=
path'.<BR>??? Undefined function or variable 'ja=
vaaddclasspath'.</P>
<P>2 jre1.4.2</P>
=
<P> matlab7=D7=D4=B4=F8=B5=C4=CA=C71.4.2=A3=BBmatlab6=D7=D4=B4=F8=B5=
=C4=CA=C71.3.=BF=C9=D2=D4=D4=DAD:\MATLAB7\sys\java\jre\win32=CF=C2=BF=B4=B5=
=BD=A1=A3=C8=E7=B9=FB=D7=B0=C1=CBmatlab7=A3=AC=CA=B9=D3=C3=CB=FC=D7=D4=B4=
=F8=B5=C41.4.2=BE=CD=BF=C9=D2=D4=C1=CB=A3=AC=D3=C8=C6=E4=B2=BB=D2=AA=CA=B9=
=D3=C31.6=A3=AC=D2=F2=CE=AA1.6=CC=AB=D0=C2=C1=CB=A3=ACmatlab=BB=B9=B2=BB=D6=
=A7=B3=D6=A1=A3=BF=C9=D2=D4=D4=DAMatlab=CF=C2=CA=B9=D3=C3 version&nb=
sp;-java=B2=E9=BF=B4JVM=B0=E6=B1=BE=A1=A3</P>
=
<P> =C8=E7=B9=FB=C4=E3=CF=EB=CA=B9=D3=C31.5=B5=C4=BB=B0=A3=ACC:\Prog=
ram Files\Java\jre1.5.0_10=A3=BB=B0=D1jre1.5.0_10=D5=E2=B8=F6=CE=C4=BC=
=FE=BC=D0=BF=BD=B1=B4=B5=BDD:\MATLAB7\sys\java\jre\win32=CF=C2=A3=AC=C8=BB=
=BA=F3=D4=F6=BC=D3=BB=B7=BE=B3=B1=E4=C1=BFMATLAB_JAVA=A3=BAD:\MATLAB7\sys=
\java\jre\win32\jre1.5.0_10=A1=A3=D5=E2=D2=BB=B2=BD=C8=E7=B9=FB=D3=D0=CE=CA=
=CC=E2=B5=C4=BB=B0=A3=AC=D6=D8=C6=F4Matlab=BB=E1=B8=F8=B3=F6=B4=ED=CE=F3=CC=
=E1=CA=BE=A1=A3=D5=D2=B2=BB=B5=BD=CA=B2=C3=B4=CA=B2=C3=B4=CE=C4=BC=FE...<=
/P>
<P>3 Weka3.4.10</P>
=
<P> =CA=B9=D3=C3weka=B0=E6=B1=BE=B5=CD=D2=BB=D0=A9=BC=B4=BF=C9=
=A3=AC=B8=DF=B5=C4=B2=BB=D0=D0=A3=AC=D2=F2=CE=AA=B8=DF=B0=E6=B1=BE=B5=C4w=
eka=BF=C9=C4=DC=CA=C7=D3=C3=B8=DF=B0=E6=B1=BE=B5=C4jvm=D6=A7=B3=D6=B5=C4=A1=
=A3</P>
=
<P>=CE=D2=CA=B9=D3=C3=B5=C4=D7=E9=BA=CF=CA=C7 matlab7=A3=A8R14=A3=A9=
+jre1.4.2=A3=A8matlab7=D7=D4=B4=F8=B5=C4=A3=AC=B2=BB=D0=E8=D2=AA=C8=CE=BA=
=CE=C9=E8=D6=C3=A3=A9+Weka3.4.10</P>
<P>=C8=FD =CA=B9=D3=C3=B7=BD=B7=A8</P>
=
<P>1 =CF=C2=D4=D8spider=A3=AC=D3=D0core=BA=CDextra=C1=BD=B8=F6=D1=B9=
=CB=F5=B0=FC=A3=AC=B0=D1=CB=FB=C3=C7=BD=E2=D1=B9=B5=BD=CD=AC=D2=BB=B8=F6=CE=
=C4=BC=FE=BC=D0spider=CF=C2=C3=E6=A3=AC=C8=BB=BA=F3=B7=C5=B5=BD$matlabroo=
t\toolbox=CF=C2=C3=E6</P>
=
<P>2=CF=C2=D4=D8weka3.4.10=A3=AC=D5=D2=B5=BDweka.jar=B7=C5=B5=BD$matlabro=
ot\java\jar=CF=C2=C3=E6</P>
=
<P>3 =C6=F4=B6=AFMatlab=B4=F2=BF=AA$matlabroot\toolbox\spider\use_sp=
ider.m=D4=CB=D0=D0</P>
=
<P>=CC=E1=CA=BEspider=B5=C4=D2=BB=D0=A9=D0=C5=CF=A2=BA=CD WEKA =
support enabled!=B1=ED=CA=BE=B3=C9=B9=A6=C1=CB=A1=A3</P>
=
<P>=C8=BB=BA=F3=BF=C9=D2=D4=CA=B9=D3=C3 help spider=C3=FC=C1=EE=
=B2=E9=BF=B4=D0=C5=CF=A2=A3=AC=CB=FB=B5=C4=B9=A6=C4=DC=C1=D0=B3=F6=C8=E7=B8=
=BD=C2=BC=A3=AC=C8=BB=BA=F3=BE=CD=BF=C9=D2=D4=D1=B5=C1=B7=C1=CB=A1=A3</P>=
=
<P>=CB=C4 =D2=BB=B8=F6=BC=F2=B5=A5=B5=C4=C0=FD=D7=D3</P>
=
<P>X=3Drand(50)-0.5; Y=3Dsign(sum(X,2));<BR>dtrain=3Ddata(X,Y);<BR>%=
=C9=FA=B3=C9=D1=B5=C1=B7=BC=AF=A3=AC=D2=B2=BF=C9=D2=D4=CA=B9=D3=C3load()=B4=
=D3=CE=C4=BC=FE=B6=C1=C8=A1</P>
=
<P>model=3Dtrain(svm,dtrain));<BR>%=CA=B9=D3=C3=BA=AF=CA=FDtrain=A3=A8=A3=
=A9=D1=B5=C1=B7=C4=A3=D0=CD</P>
=
<P>rtest=3Dtest(dtest,model);<BR>%=CA=B9=D3=C3=D1=B5=C1=B7=BA=C3=B5=C4=C4=
=A3=D0=CD=B6=D4=D1=E9=D6=A4=BC=AFdtest=B2=E2=CA=D4=A3=AC=B7=B5=BB=D8=B2=E2=
=CA=D4=BD=E1=B9=FB</P>
<P>=CE=E5 =B8=BD=C2=BCspider=D0=C5=CF=A2</P>
=
<P>=D7=EE=D0=C2spider Version 1.71 (24/7/2006)</P>
=
<P> Basic library objects. <BR> &nbs=
p;data - Storin=
g input data and output results <BR> &=
nbsp; data_global - Implementation of&nbs=
p;data object that limits memory overhead<BR>&nb=
sp; algorithm - Generic =
;algorithm object<BR> group =
- Groups sets of object=
s together (algorithms or data) <BR> =
loss -&n=
bsp;Evaluates loss functions<BR> get_mea=
n - Takes mean loss over=
groups of algs<BR> chain &nbs=
p; - Builds chains of&n=
bsp;objects: output of one to input of =
;another<BR> param &n=
bsp; - To train and test different&nb=
sp;hyperparameters of an object<BR>  =
;cv -&nb=
sp;Cross validation using objects given data<BR>=
kernel -=
Evaluates and caches kernel functions<BR> =
distance - Evaluates&=
nbsp;and caches distance functions<BR> <BR> &nbs=
p;Statistical Tests objects.<BR> wilcoxo=
n - Wilcoxon test of sta=
tistical significance of results<BR> &nbs=
p;corrt_test - Corrected resampled t-test=
- for dependent trials<BR> <BR> Data=
set objects.<BR> spiral &nbs=
p; - Spiral dataset generator.<BR> &=
nbsp; toy  =
; - Generator of dataset with only a&n=
bsp;few relevant features<BR> toy2d =
; - Simple 2d Gau=
ssian problem generator<BR> toyreg =
- Linear Regression wi=
th o outputs and n inputs <BR> <BR>&nb=
sp; Pre-Processing objects<BR> normalize=
- Simple normalization of dat=
a<BR> map  =
; - General user specified mapping&n=
bsp;function of data<BR> <BR> Density Esti=
mation objects.<BR> parzen &=
nbsp; - Parzen's windows kernel dens=
ity estimator<BR> indep &nbs=
p; - Density estimator which a=
ssumes feature independence<BR> bayes&nb=
sp; - Classifer based&=
nbsp;on density estimation for each class<BR>&nb=
sp; gauss  =
;- Normal distribution density estimator<BR> &nb=
sp; &nbs=
p; <BR> Pattern =
Recognition objects.<BR> svm  =
; - Support Vector&nbs=
p;Machine (svm)<BR> c45 &nbs=
p; - C4.5 for binary&nb=
sp;or multi-class <BR> knn &n=
bsp; - k-nearest neigh=
bours<BR> platt  =
; - Conditional Probability estimation for=
margin classifiers<BR> mksvm  =
; - Multi-Kernel LP-SVM<BR>&=
nbsp; anorm &nb=
sp;- Minimize the a-norm in alpha space&nbs=
p;using kernels<BR> lgcz &nb=
sp; - Local and Global C=
onsistent Learner <BR> bagging &nbs=
p; - Bagging Classifier<BR> &=
nbsp; adaboost - ADABoost met=
hod<BR> hmm &nb=
sp; - Hidden Markov Model <BR> =
loom &nb=
sp;- Leave One Out Machine <BR>  =
; l1 &nbs=
p;- Minimize l1 norm of w for a l=
inear separator <BR> kde &nbs=
p; - Kernel Dependency=
Estimation: general input/output machine<BR> &n=
bsp; dualperceptron -=
Kernel Perceptron<BR> ord_reg_perceptro=
n - Ordinal Regression Perceptron (S=
hen et al.)<BR> splitting_perceptron&nbs=
p;- Splitting Perceptron (Shen et al.)<BR> =
budget_perceptron - Sparse,=
online Pereceptron (Crammer et al.)<BR> &n=
bsp; randomforest - Random Forest Decision=
Trees WEKA-Req=
uired<BR> j48 &=
nbsp; - J48 Decision Trees for =
binary WEKA-Required<BR>&n=
bsp;<BR> Multi-Class and Multi-label objects.&n=
bsp;<BR> one_vs_rest - Voting =
;method of one against the rest (also =
for multi-label)<BR> one_vs_one &n=
bsp;- Voting method of one against one<BR>&=
nbsp; mc_svm -&=
nbsp;Multi-class Support Vector Machine by J.Wes=
ton<BR> c45 &nb=
sp; - C4.5 for binary or multi-=
class <BR> knn &=
nbsp; - k-nearest neighbours<BR> &n=
bsp; &nb=
sp; <BR> Feature Selection objects.<BR>&n=
bsp; feat_sel - Generi=
c object for feature selection + classifier=
<BR> r2w2_sel - =
SVM Bound-based feature selection<BR> &nb=
sp;rfe - =
Recursive Feature Elimination (also for the =
;non-linear case)<BR> l0 &nb=
sp; - Dual zero-norm&n=
bsp;minimization (Weston, Elisseeff)<BR>  =
;fsv - Pr=
imal zero-norm based feature selection (Mangasar=
ian)<BR> fisher  =
; - Fisher criterion feature selection<BR> =
mars &nb=
sp;- selection algorithm of Friedman (greedy&nbs=
p;selection)<BR> clustub &n=
bsp; - Multi-class feature selection using =
spectral clustering<BR> mutinf &nb=
sp; - Mutual Information for f=
eature selection.<BR> <BR>&=
nbsp; Regression objects.<BR> svr &=
nbsp; - Support =
Vector Regression<BR> gproc =
- Gaussian Process Regressio=
n <BR> relvm_r &=
nbsp;- Relevance vector machine <BR>  =
; multi_rr - (possibly multi-=
dimensional) ridge regression <BR> =
mrs &nbs=
p;- Multivariate Regression via Stiefel Constrai=
nts <BR> knn&nb=
sp; - k-nearest=
neighbours<BR> multi_reg &n=
bsp;- meta method for independent multiple =
output regression<BR> kmp &n=
bsp; - kernel matching =
pursuit<BR> kpls &nbs=
p; - kernel partial least squares<BR=
> lms &nb=
sp; - least mean squared regression [=
now obselete due to multi_rr]<BR> &n=
bsp;rbfnet - Radial Ba=
sis Function Network (with moving centers)<BR>&n=
bsp; reptree - ⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?