|
--- |
|
license: bsd-3-clause |
|
--- |
|
# Hong Lou Meng Fine-tuned Model for Word Alignment |
|
|
|
This repository contains a fine-tuned version of the **BERT multilingual model** (`bert-base-multilingual-cased`) on the **Hong Lou Meng** dataset for word alignment tasks. This model is fine-tuned using the [awesome-align](https://github.com/neulab/awesome-align) framework and is designed for Chinese-Vietnamese (Zh-Vn) alignment. |
|
|
|
## Model Details |
|
|
|
- **Base Model:** `bert-base-multilingual-cased` |
|
- **Fine-tuned Dataset:** Excerpts from the classic "Hong Lou Meng" novel, annotated with Chinese and Vietnamese sentence pairs. |
|
- **Alignment Task:** Fine-tuned to align word pairs in parallel texts for translation and linguistic analysis. |
|
|
|
--- |
|
|
|
## Example Usage |
|
|
|
Below is an example of how to use this model for word alignment using the `transformers` library: |
|
|
|
```python |
|
from transformers import AutoTokenizer, AutoModel |
|
import torch |
|
|
|
# Load model and tokenizer |
|
model_name = "username/zh-vn-hongloumeng-align" |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
model = AutoModel.from_pretrained(model_name) |
|
|
|
# Input sentences (Chinese and Vietnamese) |
|
source_sentence = "第一回 甄士隱夢幻識通靈 賈雨村風塵懷閨秀" |
|
target_sentence = "Hồi thứ nhất: Chân Sĩ Ẩn mộng ảo ngộ đá thiêng, Giả Vũ Thôn phong trần nhớ giai nhân." |
|
|
|
# Tokenize inputs |
|
inputs = tokenizer(source_sentence, target_sentence, return_tensors="pt", padding=True, truncation=True) |
|
|
|
# Pass through model |
|
outputs = model(**inputs) |
|
|
|
# Further processing for alignment visualization or analysis would follow |
|
print("Model outputs:", outputs) |
|
|
