[R] svm regression
Hi, I am trying to use svm for regression data. this is how my data looks like: dataTrain x y z 1 4 6 2 5 4 3 7 5 classTrain a 2 3 4 dataTest x y z 1 7 2 2 8 3 classTest a 3 4 5 building the model model-svm(dataTrain,classTrain,type=nu-regression) pred - predict(model, dataTest) pred 12 3.008842 3.120078 I don't understand what the above value means.i need the results similar to classTest. Thanks, Joji. [[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.
[R] svm regression
Hi, I am having trouble with svm regression.it is not giving the right results. example model - svm(dataTrain,classTrain,type=eps-regression) predict(model, dataTest) 36 37 38 39 40 41 42 -13.838257 -1.475401 10.502739 -3.047656 -8.713697 3.812873 1.741999 43 44 45 46 47 48 49 -6.034361 -13.469742 7.628642 -22.197060 -3.417444 -8.536890 -11.876133 50 -5.877457 My dataSet has 50 columns and 19 rows my classSet has 50 columns and 1 row My dataTrain has 35(1:35) columns and 19 rows My classTrain has 35(1:35) columns and 1 row My dataTest has 15(36:50) columns and 19 rows My classTest has 15(36:50) columns and 1 row My results should be as follows: [1] -25.70 30.30 -58.50 -1.12 7.62 -16.10 -48.50 21.10 12.60 -43.00 [11] -47.30 -47.90 -38.40 -21.30 22.40 But instead i get the wrong values.can anyone help me with it. Thanks, Joji. [[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.
Re: [R] svm regression
On 18.02.2010 17:54, madhu sankar wrote: Hi, I am trying to use svm for regression data. this is how my data looks like: dataTrain x y z 1 4 6 2 5 4 3 7 5 classTrain a 2 3 4 dataTest x y z 1 7 2 2 8 3 classTest a 3 4 5 building the model model-svm(dataTrain,classTrain,type=nu-regression) pred- predict(model, dataTest) pred 12 3.008842 3.120078 I don't understand what the above value means.i need the results similar to classTest. They are similar, aren't they? You talk about svm *regression* rather than classification and call your object class. which makes me worry you failed to understand some basics of the method. Uwe Ligges Thanks, Joji. [[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. __ 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.
Re: [R] svm regression
On 18.02.2010 19:43, madhu sankar wrote: Hi, I am having trouble with svm regression.it is not giving the right results. example model- svm(dataTrain,classTrain,type=eps-regression) predict(model, dataTest) 36 37 38 39 40 41 42 -13.838257 -1.475401 10.502739 -3.047656 -8.713697 3.812873 1.741999 43 44 45 46 47 48 49 -6.034361 -13.469742 7.628642 -22.197060 -3.417444 -8.536890 -11.876133 50 -5.877457 My dataSet has 50 columns and 19 rows my classSet has 50 columns and 1 row My dataTrain has 35(1:35) columns and 19 rows My classTrain has 35(1:35) columns and 1 row My dataTest has 15(36:50) columns and 19 rows My classTest has 15(36:50) columns and 1 row Same problems as in my last mail: I fear you are mixing up several things: regression vs. classification, rows vs. columns My results should be as follows: [1] -25.70 30.30 -58.50 -1.12 7.62 -16.10 -48.50 21.10 12.60 -43.00 [11] -47.30 -47.90 -38.40 -21.30 22.40 Why do you know? Uwe Ligges But instead i get the wrong values.can anyone help me with it. Thanks, Joji. [[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. __ 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.
Re: [R] svm regression/classification
Hi Nancy, Comments in line: On Sun, Dec 27, 2009 at 3:34 AM, Nancy Adam nancyada...@hotmail.com wrote: Hi everyone, Can anyone please tell whether there is a difference between the code for using svm in regression and code for using svm in classification? This is my code for regression, should I change it to do classification?: I'm not sure how to answer your question ... are you asking how you can explicitly tell the `svm` function to do classification vs. regression? Or are you asking if classification is better suited for your problem than regression? Are you trying to predict a label or some range of real valued numbers? Can you show us your `y` vector? -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 __ 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.
Re: [R] svm regression/classification
Hi steve, Thank you so much for your reply.Im asking about the difference between two cases:1) when I use svm in a regression system and 2) when I use svm in a classification system. Is the code of using svm in these two cases the same?This is the code for a regression system: my_svm_model - function(myformula, mydata, mytestdata) { mymodel - svm(myformula, data=mydata) mytest - predict(mymodel, mytestdata) error - mytest - mytestdata[,1] -sqrt(mean(error**2)) }Many thanks, Nancy Date: Tue, 29 Dec 2009 10:36:36 -0500 Subject: Re: [R] svm regression/classification From: mailinglist.honey...@gmail.com To: nancyada...@hotmail.com CC: r-help@r-project.org Hi Nancy, Comments in line: On Sun, Dec 27, 2009 at 3:34 AM, Nancy Adam nancyada...@hotmail.com wrote: Hi everyone, Can anyone please tell whether there is a difference between the code for using svm in regression and code for using svm in classification? This is my code for regression, should I change it to do classification?: I'm not sure how to answer your question ... are you asking how you can explicitly tell the `svm` function to do classification vs. regression? Or are you asking if classification is better suited for your problem than regression? Are you trying to predict a label or some range of real valued numbers? Can you show us your `y` vector? -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 _ Keep your friends updatedeven when youre not signed in. [[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.
Re: [R] svm regression/classification
Hi Nancy, 2009/12/30 Nancy Adam nancyada...@hotmail.com: Hi steve, Thank you so much for your reply. I’m asking about the difference between two cases: 1) when I use svm in a regression system and 2) when I use svm in a classification system. Is the code of using svm in these two cases the same? Getting the `svm` function to perform classification vs. regression can be controlled by setting its `type` parameter. The help page for the function ?svm suggests that this is automatically picked depending on what type of element your y vector is, eg. it defaults to classification if your `y` is a vector of factors. That having been said, you can set this parameter explicitly so that you're sure of what the function is doing, eg: ## classification: mymodel - svm(myformula, data=mydata, type='C-classification') ## regression mymodel - svm(myformula, data=mydata, type='eps-regression') This is the code for a regression system: my_svm_model - function(myformula, mydata, mytestdata) { mymodel - svm(myformula, data=mydata) mytest - predict(mymodel, mytestdata) error - mytest - mytestdata[,1] -sqrt(mean(error**2)) } That's not really code for a regression system -- as I said above, performing regression vs. classification depends on what type of vector your `y` labels turns out to be, given your formula (unless you explicitly set type='something'). It looks like your `my_svm_model` is a function that calculates (the negative of) the root-mean-squared-error (why negative, btw?). This performance calculation is appropriate for regression, but not for classification. For classification you probably want to report the accuracy of the labels, eg something like: mytest - predict(mymodel, mytestdata, type='C-classification') accuracy - sum(mytest == mytestdata[,1]) / length(mytest) As I said in my earlier email, it's not really appropriate to try, say, regression and report accuracy like as its defined for classification. I'm not sure if I'm answering your question, partly because I'm not really sure what you're really asking, ie. I'm not sure if you're confused as to whether or not you should be doing classification or regression, or do you know which of the two you want to do but you don't understand how to get `svm` to perform the one you want? Please clarify the above point if you still need more help. 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 __ 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.
[R] svm regression/classification
Hi everyone, Can anyone please tell whether there is a difference between the code for using svm in regression and code for using svm in classification? This is my code for regression, should I change it to do classification?: train - read.table(trainingset.txt,sep=;) test - read.table(testset.txt,sep=;) svmmodelfitness - function(myformula,mydata,mytestdata) { mymodel - svm(myformula,data=mydata) mytest - predict(mymodel, mytestdata) error - mytest - mytestdata[,1] -sqrt(mean(error**2)) } Many thanks, Nancy _ [[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.
Re: [R] SVM regression
Thank you very much! Eleni On Fri, Dec 11, 2009 at 7:19 PM, Steve Lianoglou mailinglist.honey...@gmail.com wrote: Hi Eleni, On Dec 11, 2009, at 12:04 PM, Eleni Christodoulou wrote: Dear R users, I am trying to apply SVM regression for a set of microarray data. I am using the function svm() under the package {e1071}. Can anyone tell me what the *residuals *value represents? I have some observed values *y_obs* for the parameter that I want to estimate and I would expect that *svm$residuals = y_obs - svm$fitted. *However, this does not happen...Does anyone have any idea on that? This actually is what's happening. The $residuals that are reported in the model are against your *scaled* y-vector. So, with your data: R m - svm(x,y) R all(scale(y) - predict(m,x) == m$residuals) [1] TRUE -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/contacthttp://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.
[R] SVM regression
Dear R users, I am trying to apply SVM regression for a set of microarray data. I am using the function svm() under the package {e1071}. Can anyone tell me what the *residuals *value represents? I have some observed values *y_obs* for the parameter that I want to estimate and I would expect that *svm$residuals = y_obs - svm$fitted. *However, this does not happen...Does anyone have any idea on that? Thanks a lot! Eleni C.* * [[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.
Re: [R] SVM regression
Hi Eleni, On Dec 11, 2009, at 12:04 PM, Eleni Christodoulou wrote: Dear R users, I am trying to apply SVM regression for a set of microarray data. I am using the function svm() under the package {e1071}. Can anyone tell me what the *residuals *value represents? I have some observed values *y_obs* for the parameter that I want to estimate and I would expect that *svm$residuals = y_obs - svm$fitted. *However, this does not happen...Does anyone have any idea on that? This actually is what's happening. The $residuals that are reported in the model are against your *scaled* y-vector. So, with your data: R m - svm(x,y) R all(scale(y) - predict(m,x) == m$residuals) [1] TRUE -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 __ 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.
[R] SVM regression code
Dear R user, I am looking for SVM regression in R. It willl be helpful for me if some one send me SVM regression code. Thanks Alex [[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.