VOCALINLP/portuguese_capitalization_punctuation_restoration_sanivert

Portuguese punctuation and capitalisation restoration model

Details of the model

This is a reduced version of the Portuguese capitalisation and punctuation restoration model developed by VÓCALI as part of the SANIVERT project.

You can try the model in the following SPACE

Details of the dataset

This is a neuralmind/bert-base-portuguese-cased model fine-tuned for punctuation restoration using the following data distribution.

Language	Number of text samples	Number of tokens
Portuguese	2,974,058	49,720,263

Evaluation Metrics

The metrics used to the evaluation of the model are the Macro and the Weighted F1 scores.

Funding

This work was funded by the Spanish Government, the Spanish Ministry of Economy and Digital Transformation through the Digital Transformation through the "Recovery, Transformation and Resilience Plan" and also funded by the European Union NextGenerationEU/PRTR through the research project 2021/C005/0015007

How to use the model

The metrics used to the evaluation of the model are the Macro and the Weighted F1 scores.

from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer
import torch

def get_result_text_es_pt (list_entity, text, lang):
    result_words = []
    tmp_word = ""
    if lang == "es":
        punc_tags = ['¿', '?', '¡', '!', ',', '.', ':']
    else:
        punc_tags = ['?', '!', ',', '.', ':']
    
    for idx, entity in enumerate(list_entity): 
        tag = entity["entity"]
        word = entity["word"]
        start = entity["start"]
        end = entity["end"]
        
        # check punctuation
        punc_in = next((p for p in punc_tags if p in tag), "")
                
        subword = False
        # check subwords
        if word[0] == "#": 
            subword = True
            if tmp_word == "":
                p_s = list_entity[idx-1]["start"]
                p_e = list_entity[idx-1]["end"]
                tmp_word = text[p_s:p_e] + text[start:end]
            else: 
                tmp_word = tmp_word + text[start:end]
            word = tmp_word
        else:
            tmp_word = ""
            word = text[start:end]
            
        if tag == "l": 
            word = word 
        elif tag == "u":
            word = word.capitalize()
        # case with punctuation
        else:
            if tag[-1] == "l":
                word = (punc_in + word) if punc_in in ["¿", "¡"] else (word + punc_in)
            elif tag[-1] == "u":
                word = (punc_in + word.capitalize()) if punc_in in ["¿", "¡"] else (word.capitalize() + punc_in)     
        
        if subword == True: 
            result_words[-1] = word
        else:
            result_words.append(word)

    return " ".join(result_words)

lang = "pt"
model_path = "VOCALINLP/portuguese_capitalization_punctuation_restoration_sanivert"

model = AutoModelForTokenClassification.from_pretrained(model_path)
tokenizer = AutoTokenizer.from_pretrained(model_path)

pipe = pipeline("token-classification", model=model, tokenizer=tokenizer)
text = "é preciso fazer análises ao sangue à urina e aos ouvidos"
result = pipe(text)

print("Source text: "+ text)
result_text = get_result_text_es_pt(result, text, lang)
print("Restored text: " +result_text)

Created by VOCALI SISSTEMAS INTELIGENTES S.L.