utils/summary_utils.py

import re

from langdetect import detect
from transformers import pipeline

from utils.tag_utils import filter_tags

AiSummaryVersion = 2
MinTagScore = 0.7
summarization_pipeline = pipeline("summarization", model="csebuetnlp/mT5_multilingual_XLSum")
en_translation_pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-mul-en")
classification_pipe = pipeline("text-classification", model="Yueh-Huan/news-category-classification-distilbert")
tag_gen_pipe_1 = pipeline("text-classification", model="yiyanghkust/finbert-esg-9-categories")
tag_gen_pipe_2 = pipeline("text-classification", model="dima806/news-category-classifier-distilbert")
tag_gen_pipe_3 = pipeline("text-classification", model="elozano/bert-base-cased-news-category")


def summarize(id: str, text: str):
    if text is None or len(text) < 10:
        return {
            "ver": AiSummaryVersion
        }
    summary = get_summarization(text) if len(text) > 100 else text
    translated = get_en_translation(summary)
    tags1 = get_classification(translated)
    tags2 = get_tags(translated)
    tags = filter_tags(tags1 + tags2)
    tags = sorted(list(set(tags)))

    value = {
        "id": id,
        "ver": AiSummaryVersion,
        "summary": summary,
        "tags": tags,
    }
    return value


def get_summarization(text: str):
    try:
        result = summarization_pipeline(text)
        return result[0]['summary_text'] if isinstance(result, list) else result['summary_text']
    except:
        return None


def get_en_translation(text: str):
    if text is None:
        return None
    try:
        if is_english(text):
            return text
        result = en_translation_pipe(text)
        return result[0]['translation_text'] if isinstance(result, list) else result['translation_text']
    except:
        return None


def is_english(text):
    try:
        lang = detect(text)
        return lang == 'en'
    except:
        return False


def get_tags(text: str):
    if text is None:
        return []
    try:
        print(tag_gen_pipe_1(text))
        print(tag_gen_pipe_2(text))
        print(tag_gen_pipe_3(text))
        tags1 = [tag['label'] for tag in tag_gen_pipe_1(text) if tag['score'] >= MinTagScore]
        tags2 = [tag['label'] for tag in tag_gen_pipe_2(text) if tag['score'] >= MinTagScore]
        tags3 = [tag['label'] for tag in tag_gen_pipe_3(text) if tag['score'] >= MinTagScore]
        print(tags1)
        print(tags2)
        print(tags3)
        # print(tags1, tags2, tags3)
        return []
    except:
        return []


def get_classification(text: str):
    if text is None:
        return []
    try:
        result = classification_pipe(text)
        if isinstance(result, list):
            return [tag['label'].strip() for tag in result if tag['score'] > 0.75]
        else:
            return [result['label'].strip()] if result['score'] > 0.75 else []
    except:
        return []