# from sklearn.naive_bayes import MultinomialNB
# from sklearn.preprocessing import OneHotEncoder
# X = xigua.iloc[:,:-1]
# y = xigua.iloc[:,-1]
# onehot = OneHotEncoder()
# X = onehot.fit_transform(X)
# y = np.asarray(y).reshape(-1, 1)
# clf = MultinomialNB()
# clf.fit(X, y)
# print("*********")
# print(clf.score(X, y))
#gaussian naive bayes test
import pandas as pd
from naive_bayes import GaussianNB
gender = pd.read_csv('Gender_classification.csv', header=0, encoding='utf-8')
test_sample = gender.iloc[-1, 1:]
gender_droped = gender.drop(gender.shape[0] - 1, axis=0, inplace=False)
Xtrain = gender_droped.iloc[:, 1:]
ytrain = gender_droped.人
clf = GaussianNB()
# print(clf.normal_density(x = 0, mu = 0, sigma = 1))
clf.fit(Xtrain, ytrain)
# print(clf.predict_single_instance(test_sample))
print(clf.predict(Xtrain))
print("the accuracy of training data:", clf.score(Xtrain, ytrain))
# from scipy.stats import norm
# print(norm.pdf(0 , loc = 0, scale = 1))
