dmmcutil

Documentation

Index

• Variables
• func ConfigDefault(numClusters uint16) (dmmclust.Config, error)
• type ClusterMeta
• type DMMClusterer
  • func NewDMMClusterer(numClusters uint16, clusterSize uint, docs []dmmclust.Document) (*DMMClusterer, error)
  • func (dmmc *DMMClusterer) BuildCenterDocs()
  • func (dmmc *DMMClusterer) BuildClusterMetas() error
  • func (dmmc *DMMClusterer) BuildClustersCanonical()
  • func (dmmc *DMMClusterer) BuildDocToCluserMap() (map[int]int, map[int]map[int]int, error)
  • func (dmmc *DMMClusterer) BuildScores() error
  • func (dmmc *DMMClusterer) ClusterDocCounts() (map[int]int, float64, error)
  • func (dmmc *DMMClusterer) Inflate() error
• type Documents
  • func (docs Documents) IDs() []int

Variables

var ErrNoClusters = errors.New("no clusters error")

Functions

func ConfigDefault

func ConfigDefault(numClusters uint16) (dmmclust.Config, error)

Types

type ClusterMeta

type ClusterMeta struct {
	ID          int
	DocIDs      []int
	CenterDocID int
	Cluster     dmmclust.Cluster
}

type DMMClusterer

type DMMClusterer struct {
	Config           dmmclust.Config
	Docs             []dmmclust.Document
	DocClusters      []dmmclust.Cluster // same length as `Docs`
	DocClusterScores [][]float64        // same length as `Docs`
	CenterDocs       map[int]int        // k=clusterID, v=docID
	Clusters         []dmmclust.Cluster
	ClustersMap      map[int]dmmclust.Cluster
	ClustersMetaMap  map[int]ClusterMeta
}

func NewDMMClusterer

func NewDMMClusterer(numClusters uint16, clusterSize uint, docs []dmmclust.Document) (*DMMClusterer, error)

func (*DMMClusterer) BuildCenterDocs

func (dmmc *DMMClusterer) BuildCenterDocs()

func (*DMMClusterer) BuildClusterMetas

func (dmmc *DMMClusterer) BuildClusterMetas() error

func (*DMMClusterer) BuildClustersCanonical

func (dmmc *DMMClusterer) BuildClustersCanonical()

func (*DMMClusterer) BuildDocToCluserMap

func (dmmc *DMMClusterer) BuildDocToCluserMap() (map[int]int, map[int]map[int]int, error)

BuildDocToCluserMap returns a `map[int]int` where the kesy are the input document index and the value is the clusterID.

func (*DMMClusterer) BuildScores

func (dmmc *DMMClusterer) BuildScores() error

func (*DMMClusterer) ClusterDocCounts

func (dmmc *DMMClusterer) ClusterDocCounts() (map[int]int, float64, error)

func (*DMMClusterer) Inflate

func (dmmc *DMMClusterer) Inflate() error

type Documents

type Documents []dmmclust.Document

func (Documents) IDs

func (docs Documents) IDs() []int