knn ans SVM where svm returns error

2 次查看（过去 30 天）
nkumar
nkumar 2013-5-2
评论： Greg Heath 2015-2-10
I am using knn and svm for classification
i get correct answer using knn,but if i use svm it returns wrong result
for examples
my train dataset is
datsig1 =
2551 1973
2816 2887
7070 6897
2873 2979
2588 2675
1712 2274
909 4859
792 4914
1823 7162
878 5039
885 5534
S=
2873 2979
groups=[1;1;1;1;1;1;0;0;0;0;0]
i get 0 if i classify uing svm
kindly help
cl = svmtrain(datsig1,group);
result = svmclassify(cl,S)
result =
0

请先登录，再回答此问题。

采纳的回答

Shashank Prasanna
groups=[1;1;1;1;1;1;0;0;0;0;0]
cl = svmtrain(datsig1,groups);
result = svmclassify(cl,S)
result =
1
This gave me the right results. What version of MATLAB are you using? Clear your workspace and try it from start.
Also as with all classification algorithm results are completely dependent on your training data and also options in the classification method try different kernels etc.

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by