Hi,
I would like to use stumpy for anomaly detection. Based on the tutorial here https://www.cs.ucr.edu/~eamonn/Matrix_Profile_Tutorial_Part1.pdf page 47 shows the pathological problem when using the MP discords. If two anomalies are close together they can hide each other from having a high MP value. The solution on page 50 could be an option (works by specifying the MP value to the Mth nearest neighbor instead the first NN) but I cant figure out how to do that with the stumpy API. Can you help?
Many thanks and best