conllu

Documentation

Index

• Constants
• Variables
• func ConllU2Graph(sent *Sentence, eWord, ePOS, eWPOS, eRel, eMHost, eMSuffix *util.EnumSet) nlp.LabeledDependencyGraph
• func ConllU2GraphCorpus(corpus []*Sentence, eWord, ePOS, eWPOS, eRel, eMHost, eMSuffix *util.EnumSet) []interface{}
• func ConllU2MorphGraph(sent *Sentence, ...) nlp.MorphDependencyGraph
• func ConllU2MorphGraphCorpus(corpus []*Sentence, ...) []interface{}
• func FormatFeatures(feat map[string]string) string
• func GetMorphProperties(node *transition.TaggedDepNode, eMHost, eMSuffix *util.EnumSet) string
• func Graph2ConllUCorpus(corpus []interface{}, eMHost, eMSuffix *util.EnumSet) []interface{}
• func MergeGraphAndMorph(dep Sentence, morph nlp.MorphDependencyGraph) interface{}
• func MergeGraphAndMorphCorpus(deps, morphs []interface{}) []interface{}
• func MorphGraph2ConllCorpus(corpus []interface{}) []interface{}
• func ParseInt(value string) (int, error)
• func ParseString(value string) string
• func ParseTokenRow(record []string) (string, int, error)
• func Write(writer io.Writer, sents []interface{})
• func WriteFile(filename string, sents []interface{}) error
• type Features
  • func ParseFeatures(featuresStr string) (Features, error)
  • func (f Features) MorphHost() string
  • func (f Features) MorphSuffix() string
  • func (f Features) String() string
• type Row
  • func ParseRow(record []string) (Row, error)
  • func (r Row) String() string
• type Sentence
  • func Graph2ConllU(graph nlp.LabeledDependencyGraph, eMHost, eMSuffix *util.EnumSet) Sentence
  • func MorphGraph2ConllU(graph nlp.MorphDependencyGraph) Sentence
  • func NewSentence() *Sentence
  • func ReadFile(filename string, limit int) ([]*Sentence, bool, error)
• type Sentences
  • func Read(reader io.Reader, limit int) (Sentences, bool, error)

Constants

const (
	FIELD_SEPARATOR      = '\t'
	NUM_FIELDS           = 10
	FEATURES_SEPARATOR   = "|"
	FEATURE_SEPARATOR    = "="
	FEATURE_CONCAT_DELIM = ","
)

Variables

var (
	WORD_TYPE    = "form"
	IGNORE_LEMMA bool
	STRIP_VOICE  bool
)

Functions

func ConllU2Graph

func ConllU2Graph(sent *Sentence, eWord, ePOS, eWPOS, eRel, eMHost, eMSuffix *util.EnumSet) nlp.LabeledDependencyGraph

func ConllU2GraphCorpus

func ConllU2GraphCorpus(corpus []*Sentence, eWord, ePOS, eWPOS, eRel, eMHost, eMSuffix *util.EnumSet) []interface{}

func ConllU2MorphGraph

func ConllU2MorphGraph(sent *Sentence, eWord, ePOS, eWPOS, eRel, eMFeat, eMHost, eMSuffix *util.EnumSet) nlp.MorphDependencyGraph

func ConllU2MorphGraphCorpus

func ConllU2MorphGraphCorpus(corpus []*Sentence, eWord, ePOS, eWPOS, eRel, eMFeat, eMHost, eMSuffix *util.EnumSet) []interface{}

func FormatFeatures

func FormatFeatures(feat map[string]string) string

func GetMorphProperties

func GetMorphProperties(node *transition.TaggedDepNode, eMHost, eMSuffix *util.EnumSet) string

func Graph2ConllUCorpus

func Graph2ConllUCorpus(corpus []interface{}, eMHost, eMSuffix *util.EnumSet) []interface{}

func MergeGraphAndMorph

func MergeGraphAndMorph(dep Sentence, morph nlp.MorphDependencyGraph) interface{}

func MergeGraphAndMorphCorpus

func MergeGraphAndMorphCorpus(deps, morphs []interface{}) []interface{}

func MorphGraph2ConllCorpus

func MorphGraph2ConllCorpus(corpus []interface{}) []interface{}

func ParseInt

func ParseInt(value string) (int, error)

func ParseString

func ParseString(value string) string

func ParseTokenRow

func ParseTokenRow(record []string) (string, int, error)

func Write

func Write(writer io.Writer, sents []interface{})

func WriteFile

func WriteFile(filename string, sents []interface{}) error

Types

type Features

type Features map[string]string

func ParseFeatures

func ParseFeatures(featuresStr string) (Features, error)

func (Features) MorphHost

func (f Features) MorphHost() string

func (Features) MorphSuffix

func (f Features) MorphSuffix() string

func (Features) String

func (f Features) String() string

type Row

type Row struct {
	ID      int
	Form    string
	Lemma   string
	UPosTag string
	XPosTag string
	Feats   Features
	FeatStr string
	Head    int
	DepRel  string
	Deps    []string
	Misc    string
	TokenID int
}

A Row is a single parsed row of a conll data set

func ParseRow

func ParseRow(record []string) (Row, error)

func (Row) String

func (r Row) String() string

type Sentence

type Sentence struct {
	Deps     map[int]Row
	Tokens   []string
	Mappings nlp.Mappings
}

A Sentence is a map of Rows using their ids and a set of tokens

func Graph2ConllU

func Graph2ConllU(graph nlp.LabeledDependencyGraph, eMHost, eMSuffix *util.EnumSet) Sentence

func MorphGraph2ConllU

func MorphGraph2ConllU(graph nlp.MorphDependencyGraph) Sentence

func NewSentence

func NewSentence() *Sentence

func ReadFile

func ReadFile(filename string, limit int) ([]*Sentence, bool, error)

type Sentences

type Sentences []*Sentence

func Read

func Read(reader io.Reader, limit int) (Sentences, bool, error)