机器学习算法(六) 贝叶斯分类-模拟离散数据

贝叶斯对模拟离散数据的训练，评估和预测

# import package
import random
import numpy as np 
from sklearn.naive_bayes import CategoricalNB
from sklearn.model_selection import train_test_split

# data prepare
rng = np.random.RandomState(1)
X = rng.randint(5, size = (600, 100))
y = np.array([1, 2, 3, 4, 5, 6] * 100)
data = np.c_[X, y] # combine X and y base column

random.shuffle(data) # adjust data with random order

X = data[:,:-1] # extract feature
y = data[:, -1] # extract label
print('X:\n', X)
print('y:\n', y)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)

# train
clf = CategoricalNB(alpha = 1) # strategies for 0  Laplace smooth
clf.fit(X_train, y_train)

# evaluate
acc = clf.score(X_test, y_test)
print('test acc : %.3f' % acc)

# predic 1 data
x = rng.randint(5, size=(1, 100))
print(x)
print(clf.predict_proba(x))
print(clf.predict(x))