ZamAI Pashto Multimodal Datasets
Collection
Pashto text, voice, vision, and OCR/document datasets for open AI research. • 7 items • Updated • 1
description stringclasses 1
value |
|---|
Placeholder for object detection bounding boxes |
Languages: ps
License: cc-by-4.0
Task categories: image-to-text, image-classification
Size categories: 1K<n<10K
This dataset is part of the ZamAI Pashto data collection. It is intended for image-to-text, image-classification tasks in Pashto.
from datasets import load_dataset
dataset = load_dataset("tasal9/zamai-pashto-vision")
print(dataset)
pashto_captions: load with load_dataset("tasal9/zamai-pashto-vision", "pashto_captions")scene_labels: load with load_dataset("tasal9/zamai-pashto-vision", "scene_labels")objects: load with load_dataset("tasal9/zamai-pashto-vision", "objects")@misc{zamai_pashto_data,
title = {{ZamAI Pashto Vision}},
author = {ZamAI / Yaqoob Tasal},
year = {2024},
howpublished = {\url{https://huggingface.co/datasets/tasal9/zamai-pashto-vision}}
}
This dataset is released under the cc-by-4.0 license.