Choosing a metric for your ML/AI model is as important as that metric's score.

Understanding table stakes for an AI modeling effort

Using automation to double down on an ineffective metric.

Operating on bad metrics is worse than having no metrics at all.

Some metrics are good, some are bad. But even the good ones can be bad if you're not careful.

Measure what matters

What can the world of algorithmic (electronic) trading teach us about good ML/AI practices? (Part 9 in a series)