mt5-small-thai-qg / README.md
SuperAI2-Machima's picture
Update README.md
5353a97
|
raw
history blame
2.25 kB
metadata
tags:
  - question-generation
language:
  - thai
  - th
datasets:
  - NSC2018
  - wiki-documents-nsc
  - ThaiQACorpus-DevelopmentDataset
widget:
  - text: >-
      Is this review positive or negative? Review: Best cast iron skillet you
      will every buy.
    example_title: Sentiment analysis
  - text: >-
      Barack Obama nominated Hilary Clinton as his secretary of state on Monday.
      He chose her because she had ...
    example_title: Coreference resolution
  - text: >-
      On a shelf, there are five books: a gray book, a red book, a purple book,
      a blue book, and a black book ...
    example_title: Logic puzzles
  - text: >-
      The two men running to become New York City's next mayor will face off in
      their first debate Wednesday night ...
    example_title: Reading comprehension
license: mit

Pollawat

Google's mT5

from transformers import T5Tokenizer, T5ForConditionalGeneration, T5Config

model = T5ForConditionalGeneration.from_pretrained('SuperAI2-Machima/mt5-small-thai-qg')
tokenizer = T5Tokenizer.from_pretrained('SuperAI2-Machima/mt5-small-thai-qg')

source_text = 'บุกยึดไม้เถื่อน อดีต ส.ส.บุรีรัมย์ เตรียมสร้างคฤหาสน์ทรงไทย 1 กันยายน 2550 12:00 น. ตำรวจภูธรจ.บุรีรัมย์บุกตรวจยึดไม้แปรรูปหวงห้ามกว่า 80 แผ่น'

print('Predicted Summary Text : ')
tokenized_text = tokenizer.encode(source_text, return_tensors="pt").to(device)
summary_ids = model.generate(tokenized_text,
                                        num_beams=4,
                                        no_repeat_ngram_size=2,
                                        max_length=50,
                                        early_stopping=True)
output = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
print(output)
#Predicted Summary Text : 
#answer: 80 แผ่น question: ตํารวจภูธรจ.บุรีรัมย์บุกตรวจยึดไม้แปรรูปหวงห้ามกว่ากี่แผ่น