Hi Steve,
First of all, thanks for your answer.
I did all that you suggested and now I have some questions.
1) when I use > svm.m1$coefs I have 180 values (the number of SV) of 10 and
-10, I think that it is related with C=10 because when I change for C=100 I
have 180 values of 100 and -100. I can't understand the meaning of that in
the SVR model. Do you know what it means?
2) For svm.m1$y.scale I get two values $`scaled:center` and $`scaled:scale.
The first one is the original value and the second one is the scale value
then my b will be scaled:center. Is that correct?
3) Which is the best form to put the data for SVM: in its original form
without scale and using scale=TRUE or using normalize date using
(x-min)/(max-min) like NN with scale=FALSE?
Thanks again,
Marlene.
Bellow my model:
> svm.m1 (I chose the parameters of the model using tune.svm() )
Call:
svm(formula = QQ ~ ., data = train, cost = 10, gamma = 1e-06)
Parameters:
SVM-Type: eps-regression
SVM-Kernel: radial
cost: 10
gamma: 1e-06
epsilon: 0.1
Number of Support Vectors: 180
> str(svm.m1)
List of 30
$ call : language svm(formula = QQ ~ ., data = train, cost = 10, gamma
= 1e-06)
$ type : num 3
$ kernel : num 2
$ cost : num 10
$ degree : num 3
$ gamma : num 1e-06
$ coef0 : num 0
$ nu : num 0.5
$ epsilon : num 0.1
$ sparse : logi FALSE
$ scaled : logi [1:7] TRUE TRUE TRUE TRUE TRUE TRUE ...
$ x.scale :List of 2
..$ scaled:center: Named num [1:7] 0.23 0.23 0.23 0.234 0.238 ...
.. ..- attr(*, "names")= chr [1:7] "diff1" "diff2" "diff3" "diff4" ...
..$ scaled:scale : Named num [1:7] 0.183 0.182 0.182 0.187 0.194 ...
.. ..- attr(*, "names")= chr [1:7] "diff1" "diff2" "diff3" "diff4" ...
$ y.scale :List of 2
..$ scaled:center: num 0.227
..$ scaled:scale : num 0.182
$ nclasses : int 2
$ levels : num [1:120] -0.98398 0.00101 -0.63924 0.18159 1.72474 ...
$ tot.nSV : int 180
$ nSV : int [1:2] 0 0
$ labels : int [1:2] 0 0
$ SV : num [1:180, 1:7] 1.4296 -0.9948 -0.012 -0.6508 0.1682 ...
..- attr(*, "dimnames")=List of 2
.. ..$ : chr [1:180] "1" "2" "3" "4" ...
.. ..$ : chr [1:7] "diff1" "diff2" "diff3" "diff4" ...
$ index : int [1:180] 1 2 3 4 5 6 7 8 9 10 ...
$ rho : num 0.139
$ compprob : logi FALSE
$ probA : NULL
$ probB : NULL
$ sigma : NULL
$ coefs : num [1:180, 1] -10 10 -10 10 10 -10 -10 10 10 10 ...
$ na.action: NULL
$ fitted : Named num [1:196] 0.202 0.203 0.202 0.202 0.202 ...
..- attr(*, "names")= chr [1:196] "1" "2" "3" "4" ...
$ residuals: Named num [1:196] -1.1864 -0.2018 -0.8417 -0.0206 1.5224 ...
..- attr(*, "names")= chr [1:196] "1" "2" "3" "4" ...
$ terms :Classes 'terms', 'formula' length 3 QQ ~ diff1 + diff2 + diff3
+ diff4 + diff5 + diff6 + diff7
.. ..- attr(*, "variables")= language list(QQ, diff1, diff2, diff3, diff4,
diff5, diff6, diff7)
.. ..- attr(*, "factors")= int [1:8, 1:7] 0 1 0 0 0 0 0 0 0 0 ...
.. .. ..- attr(*, "dimnames")=List of 2
.. .. .. ..$ : chr [1:8] "QQ" "diff1" "diff2" "diff3" ...
.. .. .. ..$ : chr [1:7] "diff1" "diff2" "diff3" "diff4" ...
.. ..- attr(*, "term.labels")= chr [1:7] "diff1" "diff2" "diff3" "diff4"
...
.. ..- attr(*, "order")= int [1:7] 1 1 1 1 1 1 1
.. ..- attr(*, "intercept")= num 0
.. ..- attr(*, "response")= int 1
.. ..- attr(*, ".Environment")=<environment: R_GlobalEnv>
.. ..- attr(*, "predvars")= language list(QQ, diff1, diff2, diff3, diff4,
diff5, diff6, diff7)
.. ..- attr(*, "dataClasses")= Named chr [1:8] "numeric" "numeric"
"numeric" "numeric" ...
.. .. ..- attr(*, "names")= chr [1:8] "QQ" "diff1" "diff2" "diff3" ...
- attr(*, "class")= chr [1:2] "svm.formula" "svm"
> svm.m1$coefs (some examples, I have 180 SV)
[,1]
[1,] -10
[2,] 10
[3,] -10
[4,] 10
[5,] 10
[6,] -10
[7,] -10
[8,] 10
[9,] 10
[10,] 10
> svm.m1$SV
diff1 diff2 diff3 diff4
diff5
diff6 diff7
1 1.429563809 -0.539452422 -0.376635036 3.227053246 3.765529896
0.072436459
-0.898142838
2 -0.994802571 1.429804286 -0.540826469 -0.386515451 3.087548981
3.780174219 0.087268726
3 -0.011951338 -0.998945652 1.429470704 -0.546407873 -0.394807530
3.100422296 3.821459922
4 -0.650804641 -0.014317301 -1.000562474 1.372301171 -0.548894103
-0.391030760 3.136858202
5 0.168238056 -0.654325731 -0.015413890 -0.994106648 1.300144754
-0.545519835 -0.379505172
6 1.708038328 0.166197899 -0.655760471 -0.034752129 -0.980336501
1.308349045
-0.535096474
7 -0.585281225 1.708782321 0.165196685 -0.658332568 -0.055817075
-0.978089239 1.331999124
8 -0.683566350 -0.588683840 1.708596139 0.141129535 -0.656754703
-0.051154801 -0.970752111
9 -0.028332192 -0.687146676 -0.590083898 1.644118288 0.113678155
-0.653662187 -0.037204315
10 0.659663673 -0.030727774 -0.688598758 -0.594375599 1.562091927
0.118783181
-0.644010384
> W = t(svm.m1$coefs) %*% svm.m1$SV
> W
diff1 diff2 diff3 diff4 diff5 diff6 diff7
[1,] -10.81136 60.39054 7.716997 -62.35804 92.76012 222.9277 -65.45207
model$y.scale
> svm.m1$y.scale
$`scaled:center`
[1] 0.2273616
$`scaled:scale`
[1] 0.1823783
2009/8/3 Steve Lianoglou <mailinglist.honey...@gmail.com>
> Hi Marlene,
>
>
> On Aug 3, 2009, at 8:59 AM, marlene marchena wrote:
>
> Dear R users,
>>
>> I'm running a SVR in package e1071 but I did not able to calculate the
>> parameters w and b of the regression. I don't know how to do that and if
>> it
>> is possible to do it with this package.
>>
>> Someone have some idea. Any help would be much appreciated.
>>
>
> While you're not given the W vector directly, you can recover it from the
> support vectors, no? In LaTeX form, W looks like:
>
> \vec{W} = \sum_{i=1}^N \alpha_i y_i \vec{x}_i
>
> If "model" was the object returned from your SVM: alpha_i's are the weights
> of each vector (model$coefs[i]) and x_i are the support vectors
> (model$SV[i,]).
>
> You can use some matrix multiplication to calculate this very quickly:
>
> W <- t(model$coefs) %*% model$SV
>
> By default, the x's and y's are scaled when you call svm(..) so I think, by
> definition, your y intercept (b) should be zero. I think the values in
> model$y.scale might be what you're looking for, though (not sure).
>
> Hope that helps,
> -steve
>
> --
> Steve Lianoglou
> Graduate Student: Computational Systems Biology
> | Memorial Sloan-Kettering Cancer Center
> | Weill Medical College of Cornell University
> Contact Info:
> http://cbio.mskcc.org/~lianos/contact<http://cbio.mskcc.org/%7Elianos/contact>
>
>
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
