print('DONE FILE 19')
Bernoulli.BernoulliClass(reading.train_A, reading.words_of_tweets, reading.extra_features, 8, 2, dir + '\\Bernoulli\\RFE + One-Hot.txt')
print('DONE FILE 20')
Bernoulli.BernoulliClass(reading.train_A, reading.words_of_tweets, reading.extra_features, 8, 3, dir + '\\Bernoulli\\RFE + Bigrams.txt')
print('DONE FILE 21')
'''
##############################################################################################################################################################
# Call Logistic Regression to predict irony and evaluate the outcome
##############################################################################################################################################################
'''
LogisticRegression.Logistic_Regression(reading.train_A, reading.words_of_tweets, reading.extra_features, 7, 1, dir + '\\LogisticRegression\\Univariate Selection + TF-IDF.txt')
print('DONE FILE 1')
LogisticRegression.Logistic_Regression(reading.train_A, reading.words_of_tweets, reading.extra_features, 7, 2, dir + '\\LogisticRegression\\Univariate Selection + One-Hot.txt')
print('DONE FILE 2')
LogisticRegression.Logistic_Regression(reading.train_A, reading.words_of_tweets, reading.extra_features, 7, 3, dir + '\\LogisticRegression\\Univariate Selection + Bigrams.txt')
print('DONE FILE 3')
LogisticRegression.Logistic_Regression(reading.train_A, reading.words_of_tweets, reading.extra_features, 10, 1, dir + '\\LogisticRegression\\SVD + TF-IDF.txt')
print('DONE FILE 4')
LogisticRegression.Logistic_Regression(reading.train_A, reading.words_of_tweets, reading.extra_features, 10, 2, dir + '\\LogisticRegression\\SVD + One-Hot.txt')
print('DONE FILE 5')
data_path = "E:\\github\\LogisticRegression\\data\\"
if __name__ == "__main__":
# 线性模型，最小二乘法测试*********************************************************************************
# feature = [[1, 1, 1], [1, 2, 3], [3, 7, 3]]
# label = [[3], [6], [13]]
# linear_regression = lnr.Linear_Regression()
# print(linear_regression.least_squares_method(feature, label))
# *****************************************************************************************************
# 逻辑回归模型*********************************************************************************
data_csv = pd.read_csv(data_path + "test.csv")
feature_mat = data_csv.iloc[:, 0:4]
label_mat = data_csv["label"].to_frame()
test_lr = lr.Logistic_Regression()
# 梯度下降法
grad_ascent_model = test_lr.grad_ascent(feature_mat, label_mat, 0.001, 500)
print("梯度下降法:")
print(grad_ascent_model)
print("*******************************")
# 随机梯度上升法
origin_model = test_lr.stoc_grad_ascent(feature_mat, label_mat, 0.001)
print("随机梯度上升法:")
stoc_grad_ascent_model = []
for item in origin_model:
new_list = [item]
stoc_grad_ascent_model.append(new_list)
print(stoc_grad_ascent_model)
print("*******************************")
print("小样本随机梯度上升法")
