introduce
Logistic regression is one of the most basic classification algorithms. Its model is
, where
.
its cost function
.
For the binary classification problem, the values of y are 0 and 1. Here, we
set y=1 probability. When it is greater than or equal to 0.5, we predict the result to be 1, and when it is less than 0.5, we predict the result to be 0.
Use gradient descent algorithm
Iterative formula:
where
. The derivation process is shown in the figure below.
Vectorized expression:
.