simple_knn_com_normalizer.py

# bibliotecas
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import Normalizer
import numpy as np

# remove warnings
import warnings
warnings.filterwarnings("ignore")

# dataset
iris = load_iris()

# features e target
X = iris.data
y = iris.target

# normalizando
scaler = Normalizer()
scaler.fit(X)
X = scaler.transform(X)

scores = []
for i in range(2000):
    X_train, X_test, y_train, y_test = train_test_split(X,y)
    model = KNeighborsClassifier()
    model.fit(X_train,y_train)
    accuracy = model.score(X_test,y_test)
    scores.append(accuracy)
    
print("Média: {:.2f}%".format(np.mean(scores)*100))
print("Desvio padrão: {:.2f}%".format(np.std(scores)*100))

import matplotlib.pyplot as plt
import seaborn as sns
sns.distplot(scores)
plt.yticks([])
plt.title("Acurácias do k-NN")
plt.show()