The sigmoid function is a mathematical activation function that represents Logistic Regression.
The sigmoid function is modeled by
Essentially, the sigmoid function bounds a value between 0 and 1, and is used for models to predict the probability of an output. The function is differentiable, which means we can find its slope at any point.
The sigmoid activation function is primarily used for binary classification, as it can output a probability of a certain classification, but it will fail to assist with multiclass classification. The Softmax Activation Function is a more generalized logistic activation function that can be used for multiclass classification.