Skip to content

Latest commit

 

History

History
14 lines (13 loc) · 813 Bytes

README.md

File metadata and controls

14 lines (13 loc) · 813 Bytes
Raw

Adam Gradient Descent implementation

In these short codes I implemented an example of the Adam gradient descent optimization algorithm. The optimization is done on a simple convex cost function f(x,y) = x^2 + y^2.

Usage

  1. python3 space_2d.py -> Plot the mathematical space in 2D
  2. python3 space_3d.py -> Plot the mathematical space in 3D
  3. python3 adam_implementation -> Run the Adam Gradient Optimization over the function


    plot
    plot
    plot

Credits

  1. https://machinelearningmastery.com/adam-optimization-from-scratch/, Code Adam Optimization Algorithm From Scratch, Jason Brownlee, 2021
  2. https://arxiv.org/abs/1412.6980, Adam: A Method for Stochastic Optimization, Diederik P. Kingma, Jimmy Ba, 2014