# Copyright (c) 2023-2024 Philip May
# This software is distributed under the terms of the MIT license
# which is available at https://opensource.org/licenses/MIT
"""This module offers tools for `fastText <https://fasttext.cc/docs/en/support.html>`_.
Hint:
Use pip to install the necessary dependencies for this module:
``pip install mltb2[fasttext]``
"""
import os
from dataclasses import dataclass, field
from typing import List, Optional
import fasttext
from fasttext.FastText import _FastText
from mltb2.files import fetch_remote_file, get_and_create_mltb2_data_dir
[docs]@dataclass
class FastTextLanguageIdentification:
"""Identify languages of a text."""
model: _FastText = field(init=False, repr=False)
def __post_init__(self):
"""Do post init."""
self.model = fasttext.load_model(self.get_model_path_and_download())
[docs] @staticmethod
def get_model_path_and_download() -> str:
"""Get the model path and download it if needed.
Returns:
The full path to the downloaded model file.
"""
model_filename = "lid.176.bin"
mltb2_data_home = get_and_create_mltb2_data_dir()
model_full_path = os.path.join(mltb2_data_home, model_filename)
if not os.path.exists(model_full_path):
model_url = "https://dl.fbaipublicfiles.com/fasttext/supervised-models/lid.176.bin"
sha256_checksum = "7e69ec5451bc261cc7844e49e4792a85d7f09c06789ec800fc4a44aec362764e"
fetch_remote_file_path = fetch_remote_file(
dirname=mltb2_data_home,
filename=model_filename,
url=model_url,
sha256_checksum=sha256_checksum,
)
assert fetch_remote_file_path == model_full_path
return model_full_path
[docs] def __call__(self, text: str, num_lang: int = 10, always_detect_lang: Optional[List[str]] = None):
"""Identify languages of a given text.
Args:
text: the text for which the language is to be recognized
num_lang: number of returned language probabilities
always_detect_lang: A list of languages that should always be returned
even if not detected. If the language is not detected, the probability
is set to 0.0.
Returns:
A dict from language to probability.
This dict contains no more than ``num_lang`` elements.
So it is not guaranteed that the language you want to recognize is
included in the dict. This is the case when the probability is very low.
Possible languages are: ``af als am an ar arz as ast av az azb ba bar
bcl be bg bh bn bo bpy br bs bxr ca cbk ce ceb ckb co cs cv cy da de
diq dsb dty dv el eml en eo es et eu fa fi fr frr fy ga gd gl gn gom
gu gv he hi hif hr hsb ht hu hy ia id ie ilo io is it ja jbo jv ka kk
km kn ko krc ku kv kw ky la lb lez li lmo lo lrc lt lv mai mg mhr min
mk ml mn mr mrj ms mt mwl my myv mzn nah nap nds ne new nl nn no oc or
os pa pam pfl pl pms pnb ps pt qu rm ro ru rue sa sah sc scn sco sd sh
si sk sl so sq sr su sv sw ta te tg th tk tl tr tt tyv ug uk ur uz vec
vep vi vls vo wa war wuu xal xmf yi yo yue zh``
"""
predictions = self.model.predict(text, k=num_lang)
languages = predictions[0]
probabilities = predictions[1]
lang_to_prob = {lang[9:]: prob for lang, prob in zip(languages, probabilities)}
if always_detect_lang is not None:
for lang in always_detect_lang:
if lang not in lang_to_prob:
lang_to_prob[lang] = 0.0
return lang_to_prob