svm - r support vector machine e1071 training not working -


i playing around support vector machines in r-language. using e1071 package.

as long follow manual pages or tutorial @ wikibooks everythings works. if try use own datasets examples things aren't anymore.

it seems model creation fails reason. @ least not getting levels on target column. below find example clarification.

maybe can me figure out doing wrong here. here code , data.

test dataset

target,col1,col2 0,1,2 0,2,3 0,3,4 0,4,5 0,5,6 0,1,2 0,2,3 0,3,4 0,4,5 0,5,6 0,1,2 0,2,3 0,3,4 0,4,5 1,6,7 1,7,8 1,8,9 1,9,0 1,0,10 1,6,7 1,7,8 1,8,9 1,9,0 1,0,10 1,6,7 1,7,8 1,8,9 1,9,0 1,0,10

r-script

library(e1071)  dataset <- read.csv("test.csv", header=true, sep=',')  tuned <- tune.svm(target~., data = dataset, gamma = 10^(-6:-1), cost = 10^(-1:1))  summary(tuned)  model  <- svm(target~., data = dataset, kernel="radial", gamma=0.001, cost=10)  summary(model)

output of summary(model) statement

+ summary(model)  call: svm(formula = target ~ ., data = dataset, kernel = "radial", gamma = 0.001,      cost = 10)  parameters:    svm-type:  eps-regression   svm-kernel:  radial         cost:  10        gamma:  0.001      epsilon:  0.1   number of support vectors:  28 >

wikibooks examaple

if compare output output of wikibooks example, it's missing information. please notice "levels"-section in output:

library(mass) library(e1071) data(cats) model  <- svm(sex~., data = cats) summary(model)

output

> summary(model)  call: svm(formula = sex ~ ., data = cats)  parameters:    svm-type:  c-classification   svm-kernel:  radial         cost:  1        gamma:  0.5   number of support vectors:  84   ( 39 45 )  number of classes:  2   levels:   f m

putting roland's answer in proper "answer" format:

target numeric

sex factor

let me give few more suggestions:

  • it seems if target should factor. (it has 2 levels, 0 & 1, , suspect you're trying classify either 0 or 1.) stick in dataset$target <- factor(dataset$target) somewhere.
  • right now, because target numeric, regression model being run instead of classification.
  • it's worthwhile similar check of variables before running model (especially model). in case gave, instance, it's not obvious col1 , col2 are. if either of them grouping or classification, should make them factors, too.
    • in r, many functions have multiple ways in run, depending upon data types fed them. if feed factors model, run classification. if feed numerics, regression. common in many programming languages, , called function overloading.

Comments

Post a Comment

Popular posts from this blog

curl - PHP fsockopen help required -

i trying work on cloudflare api , have found in documentation. unfortunately have no idea how can use in fsockopen or through curl. here example of post request. need know how can make request below data post using curl or fsockopen curl https://www.cloudflare.com/api_json.html -d 'a=cache_lvl' \ -d 'tkn=8afbe6dea02407989af4dd4c97bb6e25' \ -d 'email=sample@example.com' \ -d 'z=example.com' \ -d 'v=agg' here link cloudflare https://www.cloudflare.com/docs/client-api.html#s4.2 try one: $ch = curl_init("https://www.cloudflare.com/api_json.html"); curl_setopt($ch, curlopt_returntransfer,1); curl_setopt($ch, curlopt_followlocation, 1); curl_setopt($ch, curlopt_ssl_verifypeer,false); // handling of -d curl parameter here. $param = array( 'a' => 'cache_lvl', 'tkn' => '8afbe6dea02407989af4dd4c97bb6e25', 'email' => 'sample@example.com', 'z' =...
Read more

HTTP/1.0 407 Proxy Authentication Required PHP -

i working in restricted network/limited access internet. since have use proxy make web service call. here php code calling web service , response . $client = new soapclient("http: //www.xyz.com/abc/wsdls/login.wsdl", array("trace" => 1, "exceptions" => 1, 'proxy_host' => $proxy_host, 'proxy_port' => $proxy_port, 'proxy_login' => $proxy_login, 'proxy_password' => $proxy_password)); response is fatal error: uncaught soapfault exception: [wsdl] soap-error: parsing wsdl: couldn't load ' http://www.xyz.com/abc/wsdls/login.wsdl ' : failed load external entity if copy url error message , paste in browser(with same proxy settings) works. 2nd scenario is, if pass values instead of variables giving me diffarant response $client = new soapclient("http: //www.xyz.com/abc/wsdls/login.wsdl", array("trace" => 1, "exceptions" => 1, 'proxy_host' ...
Read more

c# - Resource not found error -

i have problem accessing method on homecontroller. show code of method : [httpget] public actionresult decriptidentifiantetredirige(string login_crypter, string mdp_crypter) { string loginacrypter = _globalmanager.protegemotdepasse(login_crypter); string mdpacrypter = _globalmanager.protegemotdepasse(mdp_crypter); user userapp = new models.user(login_crypter, mdp_crypter); if (userapp.authentificationvalidee(userapp.userlogin, userapp.password)) { session["name"] = userapp.userlogin; return redirecttoaction("accueil", "home"); } else { return redirecttoaction("validerauthentification", "home"); } } then in routeconfig.cs wrote route : routes.maproute( name: "authentificationapresdecryptage", url: "{controller}/{action}/{login_crypter}/{mdp_crypter}", defaults: new...
Read more