AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Change Point Detection Algorithms
The only tuning we did, i think, was to adjust how close to the edge you can get. That's largely the main configuration you can do. You can change you're stopping criterion, which is going to be based on a p test and how many trials you run. And then it computes that same number, how similar are all the points across the whole series.