ValueError: Clustering algorithm could not initialize. Consider assigning the initial clusters manually

Question

I have data containing a mixture of numeric values and categorical values. I used K-prototype to cluster them.

init = 'Huang'
n_clusters = 50
max_iter = 100

kproto = kprototypes.KPrototypes(n_clusters=n_clusters,init=init,n_init=5,verbose=verbose)

clusters = kproto.fit_predict(data_cats_matrix,categorical=categoricals_indicies)

when I run the last code I'm getting an error as follows :

ValueError: Clustering algorithm could not initialize. Consider assigning the initial clusters manually.

Answer 1

Your data might not warrant a larger number of clusters.

Run the algorithm for lesser number of k values and note the total cost at the end. If this stops decreasing, there is no need to increase k. It's called the elbow method, you can look it up.