search

openego / eTraGo

Optimization of flexibility options for transmission grids based on PyPSA
GNU Affero General Public License v3.0
30 stars 13 forks source link

Features/#608 Optimize performance of clustering #611

Closed CarlosEpia closed 1 year ago

CarlosEpia commented 1 year ago

Fixes #608

CarlosEpia commented 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.