With Kangas, data scientists can now query their large scale datasets in a way that is natural to their problem. We also made it easy to visualize, explore, and understand data interactively and intuitively.

This is a free, open-source data exploration tool from Comet.

In this video, Comet data scientist Dhruv Nair demonstrates an object detection project using Kangas.

https://github.com/comet-ml/kangas
Also check out: https://github.com/comet-ml/kangas/tree/main/examples/hugging_face/object_detection

Chapters:
0:00-3:56 – Introduction and Project Setup
3:56-15:31 – Details about the Object Detection Script
15:31-18:19 – Basics of the DataGrid UI
18:19-18:54 – Using Sorting in the DataGrid UI
18:54-22:06 – Using Group By’s in the DataGrid UI
22:06-23:24 – Running Queries in a DataGrid UI
23:24-25:03 – Analyzing Individual Images
25:03-27:47 – Advanced Query Syntax
27:47-29:56 – Querying Image Metadata
29:56-30:55 – Conclusion

About Comet:

Comet provides a self-hosted and cloud-based meta machine learning platform allowing data scientists and teams to track, compare, explain and optimize experiments and models.

Backed by thousands of users and multiple Fortune 100 companies, Comet provides insights and data to build better, more accurate AI models while improving productivity, collaboration and visibility across teams.

https://www.comet.com

source