如何在机器学习中预测 sigmoid 函数的结果

How to predict result of sigmoid function in Machine Learning

我正在上 Coursera 机器学习课程,我对 sigmoid 函数有点困惑。

我像这样实现了 sigmoid 函数:

g = 1 ./ (1+e.^(-z));

并编写了一个函数来预测结果,看起来像

p = sigmoid(X*theta) >= 0.5

问题说

"For a student with an Exam 1 score
of 45 and an Exam 2 score of 85, you should expect to see an admission
probability of 0.776" 

但我不确定如何将这两个 x 值插入到我创建的函数中。

如果 theta 是 0.218,那么 45 分和 85 分的考试成绩如何得出 0.776 的概率?有人可以解释一下吗?

谢谢

概率由sigmoid函数给出,

 p = sigmoid(X*theta)
 # Since there are two inputs, the model will have 2 weights and a bias.
 p = sigmoid(0.45*w1+0.85*w2+b)
 # The actual output is given by
 y = 0.776
 # Loss function
 loss = (p-y)^2
 # Find the weights by minimizing the loss function using gradient descent.