July 2019
Beginner to intermediate
740 pages
16h 52m
English
Since we standardized our data before clustering, we can look at the centroids, or cluster centers, to see the Z-score that the members are closest to. The centroids' location will be the average of each of the dimensions of the points in the cluster. We can grab this with the cluster_centers_ attribute of the model. The centroid of the purple cluster is located at (17.8, 19.6), which is in the (semi-major axis, period) format; remember, these are Z-scores, so these are quite far from the rest of the data. The green cluster, on the other hand, is centered at (-0.038, -0.041).
Let's build a visualization that shows us the location of the centroids with the scaled input data and the cluster ...
Read now
Unlock full access