Closed CarlosEpia closed 1 year ago
The difference in the results when using the argument "sample_weight" from kmeans.fit() is due to the parameter "init" in the method that initializes the clustering: KMeans(). When using init="k-means++", the initial cluster centers are selected from the given points, but they are different in number, in comparison with the current methodology, due to the repetition done to represent the weight of each bus.
Fixes #608