DS3 Lab

The WordScape repository contains code for the WordScape pipeline to create datasets to train document understanding models.

🛰 Deep learning for deforestation classification and forecasting in satellite imagery

Alpha version of our data-centric vision benchmark for training data correction

Repo for "TableParser: Automatic Table Parsing with Weak Supervision from Spreadsheets" at SDU@AAAI-22

Schema which is used to define the type of a machine learning data set.