ScratchData LogoScratchData
Back to deadread's profile

Sigmoid/tanh Function

DEdeadread•Created October 27, 2024
Sigmoid/tanh Function
0
0
5 views
View on Scratch

Description

The Sigmoid Function, σ (x), is used to squish all rational numbers into a value from 0 to 1. The function is commonly used in deep learning to train AI by reducing the range of numbers given to a different function. The other function assesses the AI's actions and assigns a fitness score used to tell the AI whether it should improve into this direction or try a new tactic. However, there is another variation of the Sigmoid function called the Hyperbolic Tangent, tanh (x), which squishes the numbers from -1 to 1 instead of 0 to 1. The blue line is the ordinary Sigmoid function, and the lime line is the Hyperbolic Tangent function. I have increased the size of the two lines so it is easier to see the comparison.

Project Details

Project ID1087666999
CreatedOctober 27, 2024
Last ModifiedOctober 30, 2024
SharedOctober 27, 2024
Visibilityvisible
CommentsAllowed