Views
No views yet
jefson08/kha-roberta model. The fill-mask task predicts the most likely token(s) to replace the [MASK] token in a given sentence.from transformers import pipeline, AutoTokenizer1# Initialisation
2tokenizer = AutoTokenizer.from_pretrained('jefson08/kha-roberta')
3fill_mask = pipeline(
4 "fill-mask",
5 model="jefson08/kha-roberta",
6 tokenizer=tokenizer,
7 device="cuda", # Use "cuda" for GPU or omit for CPU
8)[MASK] token for prediction:1# Predict [MASK] token
2sentence = "Nga dei u briew u ba [MASK] bha."
3predictions = fill_mask(sentence)
4
5# Display predictions
6for prediction in predictions:
7 print(f"{prediction['sequence']} (score: {prediction['score']:.4f})")"Nga dei u briew u ba [MASK] bha."1[{'score': 0.09230164438486099,
2 'token': 6086,
3 'token_str': 'mutlop',
4 'sequence': 'Nga dei u briew u ba mutlop bha.'},
5 {'score': 0.051360130310058594,
6 'token': 2059,
7 'token_str': 'stad',
8 'sequence': 'Nga dei u briew u ba stad bha.'},
9 {'score': 0.045497000217437744,
10 'token': 1864,
11 'token_str': 'khuid',
12 'sequence': 'Nga dei u briew u ba khuid bha.'},
13 {'score': 0.04180142655968666,
14 'token': 668,
15 'token_str': 'kham',
16 'sequence': 'Nga dei u briew u ba kham bha.'},
17 {'score': 0.027332570403814316,
18 'token': 2817,
19 'token_str': 'khlaiñ',
20 'sequence': 'Nga dei u briew u ba khlaiñ bha.'}]jefson08/kha-roberta model is fine-tuned for Khasi text tasks. It uses the fill-mask pipeline to predict and replace [MASK] tokens in sentences, providing insights into contextual language understanding.pip install transformers torch