debertinha-ptbr-xsmall-assin2-rte

DeBERTinha XSmall for Recognizing Textual Entailment

Labels:

• 0 : There is no entailment between premise and hypothesis.
• 1 : There is entailment between premise and hypothesis.

Full classification example

from transformers import AutoModelForSequenceClassification, AutoTokenizer, AutoConfig
import numpy as np
import torch
from scipy.special import softmax

model_name = "sagui-nlp/debertinha-ptbr-xsmall-assin2-rte"
s1 = "Os homens estão cuidadosamente colocando as malas no porta-malas de um carro."
s2 = "Os homens estão colocando bagagens dentro do porta-malas de um carro."
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
config = AutoConfig.from_pretrained(model_name)
model_input = tokenizer(*([s1], [s2]), padding=True, return_tensors="pt")
with torch.no_grad():
    output = model(**model_input)
    scores = output[0][0].detach().numpy()
    scores = softmax(scores)
    ranking = np.argsort(scores)
    ranking = ranking[::-1]
    for i in range(scores.shape[0]):
        l = config.id2label[ranking[i]]
        s = scores[ranking[i]]
        print(f"{i+1}) Label: {l} Score: {np.round(float(s), 4)}")

Citation

@misc{campiotti2023debertinha,
      title={DeBERTinha: A Multistep Approach to Adapt DebertaV3 XSmall for Brazilian Portuguese Natural Language Processing Task}, 
      author={Israel Campiotti and Matheus Rodrigues and Yuri Albuquerque and Rafael Azevedo and Alyson Andrade},
      year={2023},
      eprint={2309.16844},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}