Dataset · alonsoapp
Large dataset of table images and their HTML code
alonsoapp/TABLET-Large
Over 3.5M training examples with table images, HTML markup, and data for 17 different tasks.
What's inside
This is a large train set for the TABLET dataset, containing 3,505,311 examples across 17 different tasks. Each element is a standalone sample including a table image, its HTML representation, and associated data for a specific task.
What you can do with it
- Training VLMs and OCR models for table structure recognition and HTML translation
- Training multimodal question-answering systems for tabular documents
- Fine-tuning information extraction models from report scans and screenshots
- Size
- 1M–10M rows
- Downloads this month
- 78.9K
- Likes
- 7
How to load it
from datasets import load_dataset
ds = load_dataset("alonsoapp/TABLET-Large")