Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,56 @@
|
|
1 |
---
|
2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
+
datasets:
|
4 |
+
- mwz/ur_para
|
5 |
+
language:
|
6 |
+
- ur
|
7 |
+
tags:
|
8 |
+
- 'paraphrase '
|
9 |
---
|
10 |
+
# Urdu Paraphrasing Model
|
11 |
+
|
12 |
+
This repository contains a pretrained model for Urdu paraphrasing. The model is based on the BERT architecture and has been fine-tuned on a large dataset of Urdu paraphrases.
|
13 |
+
|
14 |
+
## Model Description
|
15 |
+
|
16 |
+
The pretrained model is based on the BERT architecture, specifically designed for paraphrasing tasks in the Urdu language. It has been trained using a large corpus of Urdu text to generate high-quality paraphrases.
|
17 |
+
|
18 |
+
## Model Details
|
19 |
+
|
20 |
+
- Model Name: Urdu-Paraphrasing-BERT
|
21 |
+
- Base Model: BERT
|
22 |
+
- Architecture: Transformer
|
23 |
+
- Language: Urdu
|
24 |
+
- Dataset: Urdu Paraphrasing Dataset mwz/ur_para
|
25 |
+
|
26 |
+
## How to Use
|
27 |
+
|
28 |
+
You can use this pretrained model for generating paraphrases for Urdu text. Here's an example of how to use the model:
|
29 |
+
|
30 |
+
```python
|
31 |
+
from transformers import pipeline
|
32 |
+
|
33 |
+
# Load the model
|
34 |
+
model = pipeline("text2text-generation", model="path_to_pretrained_model")
|
35 |
+
|
36 |
+
# Generate paraphrases
|
37 |
+
input_text = "Urdu input text for paraphrasing."
|
38 |
+
paraphrases = model(input_text, max_length=128, num_return_sequences=3)
|
39 |
+
|
40 |
+
# Print the generated paraphrases
|
41 |
+
print("Original Input Text:", input_text)
|
42 |
+
print("Generated Paraphrases:")
|
43 |
+
for paraphrase in paraphrases:
|
44 |
+
print(paraphrase["generated_text"])
|
45 |
+
```
|
46 |
+
## Training
|
47 |
+
The model was trained using the Hugging Face transformers library. The training process involved fine-tuning the base BERT model on the Urdu Paraphrasing Dataset.
|
48 |
+
|
49 |
+
## Evaluation
|
50 |
+
The model's performance was evaluated on a separate validation set using metrics such as BLEU, ROUGE, and perplexity. However, please note that the evaluation results may vary depending on the specific use case.
|
51 |
+
|
52 |
+
## Acknowledgments
|
53 |
+
- The pretrained model is based on the BERT architecture developed by Google Research.
|
54 |
+
|
55 |
+
## License
|
56 |
+
This model and the associated code are licensed under the MIT License.
|