word2vec

README

word2vec

word2vec in go lang

Thanks to these guys - https://code.google.com/p/word2vec/

Main motivations are that I used gensim with the freebase vectors skipgram file and it ended up using 27Gb of ram. I'm not particularly intestested in writting c++ so to golang. Right now the distance part uses ~11Gb of memory for the same file.

Documentation

Index

• Constants
• type Model
  • func (m *Model) Load(fn string)
  • func (m *Model) MostSimilar(seedWords []string) []WordData
• type WordData

Constants

const N int = 40

Types

type Model

type Model struct {
	M []float32
	// contains filtered or unexported fields
}

func (*Model) Load

func (m *Model) Load(fn string)

func (*Model) MostSimilar

func (m *Model) MostSimilar(seedWords []string) []WordData

type WordData

type WordData struct {
	Distance float64
	Word     string
}