pytorch/text

Ignoring UNK words

Open

#355 创建于 2018年7月22日

在 GitHub 查看
 (7 评论) (0 反应) (0 负责人)Python (822 fork)batch import
enhancementhelp wanted

仓库指标

Star
 (3,396 star)
PR 合并指标
 (30 天内没有已合并 PR)

描述

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.

贡献者指南