metadata

license: mit
task_categories:
  - text-classification
language:
  - en
  - ur
tags:
  - urdu
  - english
  - urdu poetry
pretty_name: Urdu
size_categories:
  - 1K<n<10K

English Transliteration (Roman Urdu) of Urdu Poetry Dataset

Welcome to the English Transliteration (Roman Urdu) of Urdu Poetry Dataset! This dataset provides a collection of Urdu poetry transliterated into Roman script, making the rich literary heritage of Urdu accessible to a broader audience. Each entry in the dataset includes two columns:

Title: The transliterated title of the poem in Roman Urdu.
Poem: The full text of the poem in Roman Urdu.

This dataset is ideal for researchers and developers working on transliteration models, text generation, cross-lingual studies, and cultural analysis.

Dataset Overview

The English Transliteration (Roman Urdu) of Urdu Poetry Dataset offers a diverse array of Urdu poems that have been converted into Roman script (commonly known as Roman Urdu). By providing both the title and the complete poem in a familiar script, the dataset serves as a valuable resource for:

Transliteration research: Evaluating and training models to convert text between scripts.
Text generation: Building language models capable of generating poetic text.
Comparative literary studies: Analyzing poetic forms, themes, and cultural expressions in Urdu literature.

Each record in the dataset contains:

Poem Title: The title of the poem, transliterated into Roman Urdu.
Poem Text: The complete text of the poem in Roman Urdu.

Data Format

The dataset is available in standard tabular formats such as CSV or JSON. Each entry includes the following columns:

title (string): The transliterated title of the poem.
poem (string): The full text of the poem, transliterated into Roman Urdu.

Example (CSV format):

title,poem
"Dil Ki Kitaab","Yeh ishq nahin aasan, bas itna samajh lijiye..."
"Phoolon Ka Silsila","Khushbuon ki mehfil mein, ek phool dastaan kehta hai..."
How to Load the Dataset
You can easily load the dataset using the Hugging Face Datasets library. Below is an example in Python:

python
Copy
Edit
from datasets import load_dataset

# Replace 'your-username/roman-urdu-poetry' with the actual repository name on Hugging Face Hub
dataset = load_dataset("your-username/roman-urdu-poetry")

# Accessing the train split (if applicable, otherwise adjust accordingly)
print(dataset['train'][0])
Note: Make sure you have installed the datasets library:

bash
Copy
Edit
pip install datasets
Intended Use Cases
Transliteration Research: Develop and evaluate models for transliterating text between scripts.
Text Generation: Train language models to generate poetry in Roman Urdu.
Cross-Lingual Analysis: Examine stylistic and cultural variations in poetic expression.
Literary Studies: Explore themes, structures, and narrative techniques in Urdu poetry through a transliterated lens.
License
This dataset is distributed under the MIT License. See the LICENSE file for more details.

Contributing
Contributions are welcome! If you have additional data, improvements, or suggestions, please open an issue or submit a pull request on the GitHub repository.

Citation
If you use this dataset in your research, please consider citing it as follows:

bibtex
Copy
Edit
@dataset-ReySajju742/English-Urdu-Dataset,
  author       = Sajjad Rasool,
  title        = English Transliteration (Roman Urdu) of Urdu Poetry Dataset,
  year         = 2025,
  publisher    = ReySajju742,
  url          = https://huggingface.co/datasets/ReySajju742/English-Urdu-Dataset,
}
Contact
For any questions, suggestions, or issues, please contact [email protected]