LSSVM进行分类与回归任务

LSSVM-回归

gamma = 100;
sig2 = 13;
type = 'function estimation';

%训练模型
[alpha,b] = trainlssvm({x_train,y_train,type,gamma,sig2,'RBF_kernel'});

%模型预测
y_train_hat = simlssvm({x_t, y_t, t, γ, σ², 'RBF核'}, {α, b}, x_train);
training_root_mean_square_error = sqrt(sum((y_train_hat - y_t).^2) / size(y_t)');
y_test_hat = simlssvm({x_t, y_t, t, γ, σ², 'RBF核'}, {α,b}, x_test);
testing_root_mean_square_error = sqrt(sum((y_test_hat - y_test).^2) / size(y_test)');
fprintf("训练 RMSE = %f 测试 RMSE = %f\n", training_root_mean_square_error, testing_root_mean_square_error);

**%结果可视化