Why did prateek bhaiya used y values in
kmeans.fit(X,y)
using y values in fit() will make it a supervised learning, and i guess the whole purpose is gone,
beacuse in clustering we dont have class labels (y) , and considering only X features, our algo try to finds the similiar points and group them together,
And when, y values are already provided, its just a normal classifier. then No use of kmeans…
Because this dataset was self generated so we had y labels. but in real world challenge, we wont have y labels for dataset.
So, Would not it be just :
kmeans.fit(X)