from sklearn.metrics import pairwise_distances_argmin_min
import numpy as np
from sklearn.cluster import KMeans
kmeans = KMeans(n_clusters=n_clusters)
kmeans = kmeans.fit(encoded)

n_clusters = int(np.ceil(len(encoded)**0.6))
print(n_clusters)

avg = []
for j in range(n_clusters):
    idx = np.where(kmeans.labels_ == j)[0]
    avg.append(np.mean(idx))
closest, _ = pairwise_distances_argmin_min(kmeans.cluster_centers_, encoded)
ordering = sorted(range(n_clusters), key=lambda k: avg[k])
summary = ' '.join([sentences[closest[idx]] for idx in ordering])