Untitled

from sklearn.mixture import GaussianMixture
import numpy as np

lowest_bic = np.infty
best_gmm = None
n_components_range = range(1, 20)  # try different numbers
for n in n_components_range:
    gmm = GaussianMixture(n_components=n, covariance_type='full', random_state=42)
    gmm.fit(embeddings)
    bic = gmm.bic(embeddings)
    if bic < lowest_bic:
        lowest_bic = bic
        best_gmm = gmm

cluster_labels = best_gmm.predict(embeddings)