The linear classification formula is used as the input of logistic regression, and the classification is completed when it returns, and there is also a probability value. Used for: ad click rate, judging the gender of the user, predicting whether the user will buy a given product category, and judging whether a review is positive or negative. Logistic regression is to solve the problem of binary classification.
Logistic regression formula:
Output: [0,1] interval
g(z) is the sigmoid function
sigmoid function graph:
Loss function, optimization
The principle is the same as linear regression, but because it is a classification problem, the loss function is different, and it can only be solved by gradient descent.
Log-likelihood loss function:
The complete loss function:
The smaller the value of cost loss, the higher the accuracy of the predicted category