User Tools

Site Tools


bfgs_minimization

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

bfgs_minimization [2014/07/08 11:40]
bfgs_minimization [2014/07/08 11:40]
prokop created
Line 1: Line 1:
 +====== BFGS ======
 +it call [[http://​users.eecs.northwestern.edu/​~nocedal/​lbfgsb.html|l-bfgs-b]] routine till:
 + - the convergence criteria defined in cg.optional are reached (f(i,​j)<​fmax for all i,j)
 + - the algorithm is not able to find the smaller energy in spite of f>fmax
 + - the maximum number of steps exceeds ​
 +
 +===setting of input files:===
 +==fireball.in:​==
 +   ​iquench = -5
 +   ​nstepf = 1000 ! ! Max number of opt steps, default value = 100
 +   dt = 0.5 !/only if FIRE minimization is used - last parameter in cgopt.optional >0/
 +
 +==cgopt.optional:​==
 +This routine uses the same input file as Conjiged Gradients (iquench = -4). It does not use all parameters, but they must be there.
 +   ​0.1 ​                 ! NOT USED in BFGS
 +   ​0.1 ​                 ! NOT USED in BFGS
 +   ​0.00000001 ​          ! NOT USED in BFGS (total energy tolerance )
 +   ​0.001 ​               ! fmax - force tolerance - end of minimization if fij<fmax
 +   ​10000 ​               ! NOT MORE USED IN BFGS (Max number of opt steps) - now the max. number of steps is given by parameter nstepf in firaball.in
 +   ​0 ​                   ! number of restarts in case ftot>​fmax (it seems it nave no effect)
 +   ​0 ​                   ! if >0 then dt must be defined in fireball.in. Number of FIRE downhill steps after BFGS minimization "​fails"​
  
bfgs_minimization.txt ยท Last modified: 2014/07/08 11:40 (external edit)