Quick starts
These videos show how Foundry uses foundational models to perform common machine learning tasks.
The following videos show how to use Foundry to automate data workflows:
This page describes features currently in preview. Some improvements may not yet be documented.
Summarize text
Here, Foundry uses OpenAI GPT-4 to predict labels in text and to summarize sentiment. These predictions are then sent to Annotate for review.
A transcript of this video is available.
Detect image objects
The following video shows how to use Foundry to detect objects in images. In this case, Foundry uses Amazon Rekognition and Grounding Dino to create prelabels for later review.
A transcript of this video is available.
Updated 5 months ago