Skip to content

Latest commit

 

History

History
22 lines (20 loc) · 753 Bytes

README.md

File metadata and controls

22 lines (20 loc) · 753 Bytes

gradient-descent-armijo-wolfe

Bespoke, from scratch, implementation of Armijo-Wolfe inexact line search technique to find step length for gradient descent optimisation. The library alternative is scipy.optimize.line_search

Example inputs provided to program:

  1. roll = 'M21AI619'
  2. R = last 2 digit of roll dynamically obtained
  3. FX = (x1 − R)^2 + (x2 − 2x1)^2
  4. delFX = dynamically calculated from FX
  5. epsilon = 10^-3
  6. alpha_1 = 1.0
  7. r = 0.5
  8. beta_1 = 10^-4
  9. beta_2 = 0.9
  10. X0 = [R+3,2R-2]
  11. D = dynamically calculated from delFX
  12. k = 1

Example output of program:

  1. F(x+alpha*dK): [18.99935447 37.99813236]
  2. Alpha (step length): 0.125
  3. Dk (descent direction): [-0.00101529 0.00115317]
  4. K (number of iterations): 323