Data labeling

Turning to human annotators is often the most effective and efficient way to collect labeled training data. But annotating sample data can not only be tedious, it also introduces a degree of human error that can impact model accuracy. This data labeling app showcases how you can collect several, detailed labels for a given image while streamlining the workflow for your labeling team. Labels are saved in a Baseten Postgres database so it can be used in model training.