metadata

language:
  - ja
license: cc-by-sa-3.0
library_name: transformers
tags:
  - fastText
  - embedding
pipeline_tag: feature-extraction
widget:
  - text: 海賊王におれはなる
    example_title: ワンピース

fasttext-jp-embedding

This model is experimental.

Pretrained FastText word vector for Japanese

Usage

Google Colaboratory Example

! apt install aptitude swig > /dev/null 
! aptitude install mecab libmecab-dev mecab-ipadic-utf8 git make curl xz-utils file -y > /dev/null 
! pip install transformers torch mecab-python3 torchtyping > /dev/null 
! ln -s /etc/mecabrc /usr/local/etc/mecabrc

from transformers import pipeline
import pandas as pd
import numpy as np 

text = "海賊王におれはなる"

pipeline = pipeline("feature-extraction", model="paulhindemith/fasttext-jp-embedding", revision="2022.11.13", trust_remote_code=True)
pd.DataFrame(np.array(pipeline(text)).T, columns=pipeline.tokenizer.tokenize(text))

pipeline.tokenizer.target_hinshi = ["動詞", "名詞", "形容詞"]
pd.DataFrame(np.array(pipeline(text)).T, columns=pipeline.tokenizer.tokenize(text))

License

This model utilizes the folllowing pretrained vectors. Name: fastText
Credit: https://fasttext.cc/
License: Creative Commons Attribution-Share-Alike License 3.0
Link: https://dl.fbaipublicfiles.com/fasttext/vectors-wiki/wiki.ja.vec