python机器学习——逻辑回归例子

import numpy as np
import matplotlib.pyplot as plt
from keras.models import Sequential
from keras.layers import Dense,Activation
from keras import optimizers

#1.创建数据集
user_train=np.array([[1.85,1.05],
               [1.32,2.45],
               [1.66,3.56],
               [1.33,1.56],
               [1.90,2.33],
               [2.88,2.11],
               [1.01,2.42],
               [3.66,3.34],
               [0.11,0.34],
               [0.78,0.32],
               [0.65,0.45],
               [0.56,0.99]])


user_lable=np.array([0,0,0,0,0,0,1,1,1,1,1,1])

#显示初始数据
# plt.scatter(user_train[user_lable == 0, 0], user_train[user_lable == 0, 1])
# plt.scatter(user_train[user_lable == 1, 0], user_train[user_lable == 1, 1])
# plt.show()

#2.创建模型
model = Sequential()
model.add(Dense(input_dim=2,units=1))
model.add(Activation('sigmoid'))  #设置激活函数,将结果转化为概率
sgd = optimizers.SGD()
model.compile(loss='binary_crossentropy', optimizer=sgd)  #选择损失函数和优化器


#3.训练
print('Training')
for step in range(1000):
    const=model.train_on_batch(user_train,user_lable)
    if step%50==0:
        print('After %d training,the const:%f' %(step,const))

#4.测试
print('Testing......')
cost=model.evaluate(user_train,user_lable,batch_size=40)
print('cost= ',cost)

y_pred=model.predict(user_train)

y_pred=(y_pred*2).astype('int')
print(y_pred)

运行结果：