import pandas as pd
from sklearn.preprocessing import MinMaxScaler
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
from sklearn.metrics import silhouette_score

df = pd.read_csv('dogs.csv')

print(df.head())

features = df.columns[1:]

scaler = MinMaxScaler().fit(df[features])
x = pd.DataFrame(scaler.transform(df[features]))
x.columns = features

colors = ['red', 'green', 'blue', 'gold']

fig = plt.figure(figsize=(5,5))
plt_ind=1

for i in [3,4]:
    est = KMeans(n_clusters=i, init='random')
    est.fit(x)
    df['labels'] = est.labels_

    centers = pd.DataFrame(scaler.inverse_transform(est.cluster_centers_), columns=features)
    print('centers', centers)

    sp =fig.add_subplot(2,1,plt_ind)

    for j in range(0,i):
        cluster = df.loc[df['labels'] ==j]
        plt.scatter(cluster['height'], cluster['weight'], color = colors[j], label ="cluster %d"%j,
                    )

    sp.scatter(centers['height'], centers['weight'], color='black', marker='x', label = 'centroidi')
    plt.title('Senka %0.3f' % silhouette_score(x, est.labels_))
    plt.legend()


    plt_ind+=1

plt.tight_layout()
plt.show()