import scipy
from sklearn.datasets import load_iris

from sklearn.model_selection import train_test_split
from sklearn.cluster import KMeans
from sklearn.metrics import accuracy_score

def find_permutation(n_clusters, real_labels, labels):
    permutation=[]
    for i in range(n_clusters):
        idx = labels == i
        # Choose the most common label among data points in the cluster
        new_label=scipy.stats.mode(real_labels[idx])[0][0]
        permutation.append(new_label)
    return permutation

def plant_clustering():

    X , y  = load_iris(return_X_y=True)

    cluss = KMeans( 3 , random_state  = 0 )
    cluss.fit( X )

    y_pred = cluss.predict( X )
    #print( y_pred )
    #print( y )

    perm = find_permutation( 3 , y , y_pred )
    #print(perm )
    acc = accuracy_score( y , [ perm[label] for label in cluss.labels_])

    #print(acc)
    return acc

def main():
    print(plant_clustering())

if __name__ == "__main__":
    main()