Implement online learning for k-means

Author: mpraski

clusters.go

@@ -20,21 +20,21 @@ type Clusterer interface {
 }
 
 type HardClusterer interface {
-	Clusters() ([]HardCluster, error)
+	Guesses() []HardCluster
 
 	Predict(observation []float64) (HardCluster, error)
 
-	Online(observations chan []float64, done chan bool) chan []HardCluster
+	Online(observations chan []float64, done chan struct{}) chan []HardCluster
 
 	Clusterer
 }
 
 type SoftClusterer interface {
-	Clusters() ([]*SoftCluster, error)
+	Guesses() []*SoftCluster
 
 	Predict(observation []float64) (*SoftCluster, error)
 
-	Online(observations chan []float64, done chan bool) chan []*SoftCluster
+	Online(observations chan []float64, done chan struct{}) chan []*SoftCluster
 
 	Clusterer
 }

errors.go

@@ -3,9 +3,8 @@ package clusters
 import "errors"
 
 var (
-	ErrDimensionMismatch = errors.New("Vectors have different dimension")
-	ErrEmptySet          = errors.New("Empty training set")
-	ErrEmptyClusters     = errors.New("Empty clusters")
-	ErrZeroIterations    = errors.New("Number of iterations cannot be less than 1")
-	ErrZeroClusters      = errors.New("Number of clusters cannot be less than 1")
+	ErrEmptySet       = errors.New("Empty training set")
+	ErrNotTrained     = errors.New("You need to train the algorithm first")
+	ErrZeroIterations = errors.New("Number of iterations cannot be less than 1")
+	ErrZeroClusters   = errors.New("Number of clusters cannot be less than 1")
 )