0

I have a fairly large 1-dimensional array that I am trying to cluster. I came across several other questions on this site where the top answer is to use a Kernel Density Estimation and then locate local minima: Determine different clusters of 1d data from database and Alternative to Otsu for dividing data into two groups.

I have successfully used SciPy's gaussian_kde function to generate KDE's for my data, and they plot beautifully, but how do I calculate the local minima?

Mike
  • 1
  • 1

1 Answers1

1

Compare each value to the two neighboring values.

If it is smaller than both of them, it is a local minimum.

Has QUIT--Anony-Mousse
  • 39,639
  • 7
  • 61
  • 96