gabriellm1

enc = OrdinalEncoder()
labels = enc.fit_transform(labels[['Class']])
labels *= 100
plt.scatter(df_principal['P1'], df_principal['P2'], c=labels , cmap=plt.cm.cool)
plt.show()

gabriellm1

from sklearn.preprocessing import OrdinalEncoder

labels = pd.read_csv('labels.csv')
labels

gabriellm1

from sklearn.cluster import DBSCAN 

db_default = DBSCAN(eps = 0.008, min_samples = 10).fit(df_principal) 
labels = db_default.labels_ 

colours = {} 
colours[0] = 'r'
colours[1] = 'g'
colours[2] = 'b'
colours[3] = 'y'

gabriellm1

from sklearn.decomposition import PCA 

pca = PCA(n_components = 2) 
df_principal = pca.fit_transform(df_prepared) 
df_principal = pd.DataFrame(df_principal) 
df_principal.columns = ['P1', 'P2'] 
df_principal

gabriellm1

from sklearn.pipeline import Pipeline

num_pipeline = Pipeline(
    [
      ("imputer", SimpleImputer(strategy="median")),
    ]
)
df_prepared = num_pipeline.fit_transform(df)
df_prepared = normalize(df_prepared)

gabriellm1

from sklearn import cluster, datasets 

n_samples = 1000
noisy_moons, _ = datasets.make_moons(n_samples=n_samples, noise=.05) 
noisy_circles, _ = datasets.make_circles(n_samples=n_samples, factor=.5, noise=.05)
blobs1, _ = datasets.make_blobs(n_samples=n_samples, random_state=10, center_box=(-2, 2),cluster_std=0.2)
blobs2, _ = datasets.make_blobs(n_samples=n_samples, random_state=4, center_box=(-2, 2),cluster_std=0.4)
blobs3, _ = datasets.make_blobs(n_samples=n_samples, random_state=7, center_box=(-2, 2),cluster_std=0.3)
no_structure, _ = np.random.rand(n_samples, 2), None