func (v *ArrayVector) ToString() []byte {
sb := util.StringBuilder{}
for _, value := range v.data {
sb.Float(value)
sb.Write("|")
}
return sb.Bytes()
}
func (algo *LinearRegression) SaveModel(path string) {
sb := util.StringBuilder{}
for f, g := range algo.Model {
sb.Int64(f)
sb.Write("\t")
sb.Float(g)
sb.Write("\n")
}
sb.WriteToFile(path)
}
func (self *LinearSVM) SaveModel(path string) {
sb := util.StringBuilder{}
for f, g := range self.w.Data {
sb.Int64(f)
sb.Write("\t")
sb.Float(g)
sb.Write("\n")
}
sb.WriteToFile(path)
}
func (lr *LROWLQN) SaveModel(path string) {
sb := util.StringBuilder{}
for key, val := range lr.Model.Data {
sb.Int64(key)
sb.Write("\t")
sb.Float(val)
sb.Write("\n")
}
sb.WriteToFile(path)
}
func (v *Vector) ToString() []byte {
sb := util.StringBuilder{}
for key, value := range v.Data {
sb.Int64(key)
sb.Write(":")
sb.Float(value)
sb.Write("|")
}
return sb.Bytes()
}
func (s *Sample) ToString(includePrediction bool) []byte {
sb := util.StringBuilder{}
sb.Int(s.Label)
sb.Write(" ")
if includePrediction {
sb.Float(s.Prediction)
sb.Write(" ")
}
for _, feature := range s.Features {
sb.Int64(feature.Id)
sb.Write(":")
sb.Float(feature.Value)
sb.Write(" ")
}
return sb.Bytes()
}
func (algo *EPLogisticRegression) SaveModel(path string) {
sb := util.StringBuilder{}
for f, g := range algo.Model {
sb.Int64(f)
sb.Write("\t")
sb.Float(g.Mean)
sb.Write("\t")
sb.Float(g.Vari)
sb.Write("\n")
}
sb.WriteToFile(path)
}
func (t *Tree) ToString() []byte {
sb := util.StringBuilder{}
sb.Int(len(t.nodes))
sb.Write("\n")
for i, node := range t.nodes {
sb.Int(i)
sb.Write("\t")
sb.Int(node.left)
sb.Write("\t")
sb.Int(node.right)
sb.Write("\t")
sb.Int(node.depth)
sb.Write("\t")
sb.WriteBytes(node.prediction.ToString())
sb.Write("\t")
sb.Int(node.sample_count)
sb.Write("\t")
sb.Int64(node.feature_split.Id)
sb.Write("\t")
sb.Float(node.feature_split.Value)
sb.Write("\n")
}
return sb.Bytes()
}