metadata
license: mit
task_categories:
- text-classification
- table-question-answering
- question-answering
- zero-shot-classification
- translation
language:
- ps
pretty_name: Pashto
size_categories:
- 10K<n<100K
ZamAI-Pashto-High-Quality-Dataset
Description
This is a high-quality Pashto dataset created for enabling language-sensitive AI applications, particularly for ZamAI Pashto Keyboard and other NLP tasks.
Dataset Details
- Language: Pashto
- Use Case: Language modeling, text generation, translation, and keyboard optimization
- Source: Curated from Pashto-sensitive keyboard usage and contributions.
File Structure
The dataset includes the following files: tutor_chat_dataset.jsonl, islamic_content_dataset.jsonl, zamai_training_dataset.jsonl, zamai_training_dataset.json, README.md, cultural_content_dataset.jsonl, general_qa_dataset.jsonl, language_learning_dataset.jsonl, dataset_info.json, zamai_training_dataset.csv, business_automation_dataset.jsonl, news_summarization_dataset.jsonl
Licensing and Usage
Please ensure appropriate attribution when using this dataset in your projects.
Contact
For questions or collaborations, contact tasal9.