MCPcopy Create free account
hub / github.com/Xtra-Computing/thundersvm / load_dataset_and_train

Method load_dataset_and_train

src/test/test_svr.cpp:18–32  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

16 vector<float_type> predict_y;
17
18 float
19 load_dataset_and_train(string train_filename, string test_filename, float_type C, float_type gamma) {
20 train_dataset.load_from_file(train_filename);
21 test_dataset.load_from_file(test_filename);
22 param.gamma = gamma;
23 param.C = C;
24 param.kernel_type = SvmParam::RBF;
25 std::shared_ptr<SvmModel> model;
26 model.reset(new SVR());
27 model->train(train_dataset, param);
28 std::shared_ptr<Metric> metric;
29 metric.reset(new MSE());
30 predict_y = model->predict(test_dataset.instances(), 100);
31 return metric->score(predict_y, test_dataset.y());
32 }
33};
34
35TEST_F(SVRTest, test_set) {

Callers

nothing calls this directly

Calls 5

instancesMethod · 0.80
scoreMethod · 0.80
load_from_fileMethod · 0.45
trainMethod · 0.45
predictMethod · 0.45

Tested by

no test coverage detected