Project description

rs-bytepiece

Install

pip install rs_bytepiece

Usage

from rs_bytepiece import Tokenizer

tokenizer = Tokenizer()
# a custom model
tokenizer = Tokenizer("/path/to/model")
ids = tokenizer.encode("今天天气不错")
text = tokenizer.decode(ids)

Performance

The performance is a bit faster than the original implementation. I've tested (on my M2 16G) the《鲁迅全集》which has 625890 chars. The time unit is millisecond.

length	jieba	aho_py	aho_cy	aho_rs
100	17062.12	1404.37	564.31	112.94
1000	17104.38	1424.6	573.32	113.18
10000	17432.58	1429.0	574.93	110.03
100000	17228.17	1401.01	574.5	110.44
625890	17305.95	1419.79	567.78	108.54

Project details

These details have not been verified by PyPI

Project links

Intended Audience
Operating System
- OS Independent
Programming Language
- Python :: 3
- Rust
Topic
- Scientific/Engineering :: Artificial Intelligence

Release history Release notifications | RSS feed

This version

0.2.2

Nov 12, 2023

0.2.1

Oct 17, 2023

0.2.0

Oct 16, 2023

0.1.0

Sep 20, 2023

0.0.3

Sep 19, 2023

0.0.2

Sep 19, 2023

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rs_bytepiece-0.2.2.tar.gz (1.2 MB view hashes)

Uploaded Nov 12, 2023 Source

Built Distributions

rs_bytepiece-0.2.2-cp37-abi3-win_amd64.whl (3.8 MB view hashes)

Uploaded Nov 12, 2023 CPython 3.7+ Windows x86-64

rs_bytepiece-0.2.2-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.5 MB view hashes)

Uploaded Nov 12, 2023 CPython 3.7+ manylinux: glibc 2.17+ x86-64

rs_bytepiece-0.2.2-cp37-abi3-macosx_10_7_x86_64.whl (2.3 MB view hashes)

Uploaded Nov 12, 2023 CPython 3.7+ macOS 10.7+ x86-64

Hashes for rs_bytepiece-0.2.2.tar.gz

Hashes for rs_bytepiece-0.2.2.tar.gz
Algorithm	Hash digest
SHA256	`f86206004808f118d7581fe314387c6d0c45566cd34fda346c439c3392e912cd`
MD5	`5b7c4bd2fb2e597f4088d5b1a33bfb9f`
BLAKE2b-256	`8a22b6bbac87677550e256e049e97b2cc4a75aa33bd12bb81ac7c4709d407178`

Hashes for rs_bytepiece-0.2.2-cp37-abi3-win_amd64.whl

Hashes for rs_bytepiece-0.2.2-cp37-abi3-win_amd64.whl
Algorithm	Hash digest
SHA256	`2e6a9f8b78bc24d4856b00240c82392ebe6008b110ba482e22c00f48f81b4e6c`
MD5	`a20df1d801d72386db90a6e23dfc9efe`
BLAKE2b-256	`6230dea969abe55a7cf936c69b32b5ddda7eb09687300be9fa8923535da0d9f6`

Hashes for rs_bytepiece-0.2.2-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for rs_bytepiece-0.2.2-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`2405a38e0a03985fabeb025eae65fcafcf5aae7c5f9b065c10cc17b72ae5e723`
MD5	`52a51aebe3216eadb80bafb8150db11a`
BLAKE2b-256	`14610a22cb90c845829383640d2a9ef17ab0286e9cef9189b7346194b5df2b72`

Hashes for rs_bytepiece-0.2.2-cp37-abi3-macosx_10_7_x86_64.whl

Hashes for rs_bytepiece-0.2.2-cp37-abi3-macosx_10_7_x86_64.whl
Algorithm	Hash digest
SHA256	`14338a8c8573df2ac4dc83567a58722d18984f678ce3426237d58e689288df1d`
MD5	`485620da2afd64f4758720e3ab7704aa`
BLAKE2b-256	`4863809ac38242cf82098144777c8f82a94290d15aa04f9c45dbb57da7a78be8`