Evaluating models without test data (Practical AI #194)
Changelog Master Feed
00:00
How Did You Come Up With the Weight Watcher Tool?
The Weight Watcher tool is an open source diagnostic tool for analyzing neural networks without the need for access to training or even test data. It's been developed by Charles Wojciechowski, who previously worked as a postdoc in theoretical physics. He says he realized that nobody in the machine learning or AI community really knows about this stuff. The project has taken seven years now of research and development.
Transcript
Play full episode
