# **Khasi Fill-Mask Model**

This project demonstrates how to use the Hugging Face Transformers library to perform a fill-mask task using the **`jefson08/kha-roberta`** model. The fill-mask task predicts the most likely token(s) to replace the `[MASK]` token in a given sentence.

---

## **Usage**

### **1. Import Dependencies**

```python
from transformers import pipeline, AutoTokenizer
```

### **2. Initialize the Model and Tokenizer**

Load the tokenizer and model pipeline:

```python
# Initialisation
tokenizer = AutoTokenizer.from_pretrained('jefson08/kha-roberta')
fill_mask = pipeline(
    "fill-mask",
    model="jefson08/kha-roberta",
    tokenizer=tokenizer,
    device="cuda",  # Use "cuda" for GPU or omit for CPU
)
```

### **3. Predict the [MASK] Token**

Provide a sentence with a `[MASK]` token for prediction:

```python
# Predict [MASK] token
sentence = "Nga dei u briew u ba [MASK] bha."
predictions = fill_mask(sentence)

# Display predictions
for prediction in predictions:
    print(f"{prediction['sequence']} (score: {prediction['score']:.4f})")
```

---

## **Example Output**

Given the input sentence:

```plaintext
"Nga dei u briew u ba [MASK] bha."
```

The model might output:

```plaintext
[{'score': 0.09230164438486099,
  'token': 6086,
  'token_str': 'mutlop',
  'sequence': 'Nga dei u briew u ba  mutlop bha.'},
 {'score': 0.051360130310058594,
  'token': 2059,
  'token_str': 'stad',
  'sequence': 'Nga dei u briew u ba  stad bha.'},
 {'score': 0.045497000217437744,
  'token': 1864,
  'token_str': 'khuid',
  'sequence': 'Nga dei u briew u ba  khuid bha.'},
 {'score': 0.04180142655968666,
  'token': 668,
  'token_str': 'kham',
  'sequence': 'Nga dei u briew u ba  kham bha.'},
 {'score': 0.027332570403814316,
  'token': 2817,
  'token_str': 'khlaiñ',
  'sequence': 'Nga dei u briew u ba  khlaiñ bha.'}]
```

---

## **Model Information**

The `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.


---

## **Dependencies**

- [Transformers](https://huggingface.co/docs/transformers): Provides the pipeline and model-loading utilities.
- [PyTorch](https://pytorch.org/): Backend framework for running the model.

Install the dependencies with:

```bash
pip install transformers torch
```

---

## **Acknowledgements**

- Hugging Face [Transformers](https://huggingface.co/docs/transformers) library.
- Model by [N Donald Jefferson Thabah](https://huggingface.co/jefson08/kha-roberta).

---

## **License**

This project is licensed under the MIT License. See the [LICENSE](./LICENSE) file for more details.

---