一层隐藏层的神经网络

神经网络模型：
一、初始化参数：
W1, b1, W2, b2
二、前向计算求出cache：
Z1 = np.dot(W1, X) + b1
A1 = np.tanh(Z1)
Z2 = np.dot(W2, A1) + b2
A2 = sigmoid(Z2)
三、计算损失值：
cost = - np.sum(np.multiply(np.log(A2), Y) + np.multiply(np.log(1. - A2), 1. - Y)) / m
四、后向计算求值：
dZ2 = A2 - Y
dW2 = np.dot(dZ2, A1.T) / m
db2 = np.sum(dZ2, axis = 1, keepdims = True) / m
dZ1 = np.dot(W2.T, dZ2) * (1 - A1**2)
dW1 = np.dot(dZ1, X.T) / m
db1 = np.sum(dZ1, axis = 1, keepdims = True) / m
五、根据梯度下降方向不断迭代，优化参数，减少损失函数的值：
W1 = W1 - learning_rate * dW1
b1 = b1 - learning_rate * db1
W2 = W2 - learning_rate * dW2
b2 = b2 - learning_rate * db2