pytorch/text

Ignoring UNK words

Open

#355 aberto em 22 de jul. de 2018

Ver no GitHub
 (7 comments) (0 reactions) (0 assignees)Python (822 forks)batch import
enhancementhelp wanted

Métricas do repositório

Stars
 (3.396 stars)
Métricas de merge de PR
 (Nenhuma PRs mesclada em 30d)

Description

Cannot find the way to ignore UNK words when numericalising, i.e. instead by substituting them by a 0, it just ignore that word.

Is that implemented?

This is useful in classification problems, when you just want to remove 'UNK' words.

Guia do colaborador