scale the input data for correct learning
[scaled_instance,scaled_parameters] = svmscale(instance); [scaled_instance,scaled_parameters] = svmscale(instance,[lower,upper]); [scaled_instance] = svmscale(instance,scaled_parameters); [scaled_instance,scaled_parameters,scaled_label,scaled_label_parameters] = svmscale(instance,[lower,upper],label,[label_lower, label_upper]);
Scale your data. For example, scale each attribute to [0,1] or [-1,+1].
[label,instance]=libsvmread("demos/heart_scale"); [scaled_instance,scaled_parameters] = svmscale(instance,[-1,1]); cc = svmtrain(label,scaled_instance); [predicted_label]=svmtrain(label,svmscale(instance,scaled_parameters));