Package: wnpp Severity: wishlist Owner: Kohei Sendai <kouhei.sen...@gmail.com> X-Debbugs-Cc: debian-de...@lists.debian.org, kouhei.sen...@gmail.com
* Package name : tokenizers Version : 0.20.3 Upstream Contact: Nicolas Patry <noc...@huggingface.co> * URL : https://github.com/huggingface/tokenizers * License : Apache-2.0 Programming Lang: Python Description : Provides an implementation of today's most used tokenizers, with a focus on performance and versatility. Train new vocabularies and tokenize, using today's most used tokenizers. Extremely fast (both training and tokenization), thanks to the Rust implementation. Takes less than 20 seconds to tokenize a GB of text on a server's CPU. Easy to use, but also extremely versatile. Designed for research and production. Normalization comes with alignments tracking. It's always possible to get the part of the original sentence that corresponds to a given token. Does all the pre-processing: Truncate, Pad, add the special tokens your model needs. - Important package for vllm and used in many place with llm.
