epfl-dlab / epfl-dlab/transformers-CFG

remove legacy TokenTrie

Open
#67 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
139
Forks
23
PR merge metrics
No merged PRs in 30d

Description

Currently there is ByteTrie at https://github.com/epfl-dlab/transformers-CFG/blob/main/transformers_cfg/tokenization/trie.py#L21
and also a TokenTrie at https://github.com/epfl-dlab/transformers-CFG/blob/main/transformers_cfg/tokenization/vocab_struct.py#L13

TokenTrie is the legacy code from previous implementation and it should be completely removed and replaced by ByteTrie.

There is still dependency on it in the parsing, so may need to treat it carefully

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.