Uploaded October 2020 | Updated September 2026, 4 hours ago
When utilizing machine learning models in real world contexts, it is important to quantify how model confidence relates to real world outcomes, even in the light of limited information. This demonstration shows how to use probabilistic programming in Tensorflow to develop a statistical framework and intuition for model performance, and explores how these insights can be used to guide business logic.
When utilizing machine learning models in real world contexts, it is important to quantify how model confidence relates to real world outcomes, even in the light of limited information. This demonstration shows how to use probabilistic programming in Tensorflow to develop a statistical framework and intuition for model performance, and explores how these insights can be used to guide business logic.
