// Train using sampler.
func BenchmarkTrain2(b *testing.B) {
dim := 8
numSamp := 2000000
// Use a FloatObserver.
fs, err := model.NewFloatObserver(make([][]float64, numSamp, numSamp),
make([]model.SimpleLabel, numSamp, numSamp))
if err != nil {
b.Fatal(err)
}
mean := []float64{0.1, 0.2, 0.3, 0.4, 1, 1, 1, 1}
std := []float64{0.5, 0.5, 0.5, 0.5, 0.1, 0.2, 0.3, 0.4}
g := NewModel(dim, Name("test training"))
r := rand.New(rand.NewSource(33))
for i := 0; i < numSamp; i++ {
rv := model.RandNormalVector(r, mean, std)
fs.Values[i] = rv
}
for i := 0; i < b.N; i++ {
g.Update(fs, model.NoWeight)
g.Estimate()
g.Clear()
}
}
func TestTrainGaussian2(t *testing.T) {
if testing.Short() {
t.Skip("skipping test in short mode.")
}
dim := 8
numSamp := 2000000
// Use a FloatObserver.
values := make([][]float64, numSamp, numSamp)
labels := make([]model.SimpleLabel, numSamp, numSamp)
mean := []float64{0.1, 0.2, 0.3, 0.4, 1, 1, 1, 1}
std := []float64{0.5, 0.5, 0.5, 0.5, 0.1, 0.2, 0.3, 0.4}
g := NewModel(dim, Name("test training"))
r := rand.New(rand.NewSource(33))
for i := 0; i < numSamp; i++ {
rv := model.RandNormalVector(r, mean, std)
values[i] = rv
}
fo, err := model.NewFloatObserver(values, labels)
if err != nil {
t.Fatal(err)
}
g.Update(fo, model.NoWeight)
g.Estimate()
t.Logf("Mean: \n%+v", g.Mean)
t.Logf("STD: \n%+v", g.StdDev)
for i, _ := range mean {
if !gjoa.Comparef64(mean[i], g.Mean[i], tolerance) {
t.Errorf("Wrong Mean[%d]. Expected: [%f], Got: [%f]",
i, mean[i], g.Mean[i])
}
if !gjoa.Comparef64(std[i], g.StdDev[i], tolerance) {
t.Errorf("Wrong STD[%d]. Expected: [%f], Got: [%f]",
i, std[i], g.StdDev[i])
}
}
}
