How to optimize the output of my python program

Question

I have a python program that takes a bunch of values and calculates the spin of a particle as output. The program is not a function, so there are no inputs. I have been told to optimize three different values individually (only changing one value, while leaving the other two constant) to produce an optimized output of the y-component of spin of this particle. I think what I need to do it turn this program into a function, with the value I want to optimize being the input. Then I think I need to make a new function to iterate through all possible values of the input to produce my desired output. For example, I need to optimize the length of a dipole magnet. The y-component of spin should have a value of -1 in real life, but I need it to be accurate to 10 decimal places (-.9999999999). How to I optimize the value of the length of the dipole magnet to produce the value of y-spin to be accurate to 10 decimal places?

can you give some more information on your code? Is it a linear script? If so, you can quickly make it into a function. If you have the function, you can look into to scipy.optimize find the optimum scipy.optimize: docs.scipy.org/doc/scipy-0.14.0/reference/optimize.html — tvandenbrande
– tvandenbrande, Commented Nov 2, 2014 at 21:36
@tvandenbrande I'm sorry I'm really not very familiar with programming, so I do not know if it is a linear script. What does that mean? — Spencer
– Spencer, Commented Nov 2, 2014 at 21:57
do you use classes in python or not? If not, then you are probably using a linear script. Can you put your code here so we can look at it? — tvandenbrande
– tvandenbrande, Commented Nov 2, 2014 at 21:58

tvandenbrande · Accepted Answer · 2014-11-02 23:09:01Z

1

If I'm correct you calculate S[1], S[2] and S[3] right? Try to rewrite your code so you get something like:

def getSpin(L1,L2,d):
   ...all your calculations will be done here..
   # return the value of S[2] minus the goal value, optimization can find 0
   return S[2]-1.0

To optimise the result, use the scipy.optimize-package and evaluate on S[2]. To do so, be sure that scipy is installed on your machine. The code below is a first suggestion from my side and thus not yet tested/debugged, so please check out the documentation of the solver. In this code x0 and the bounds needs to be filled in. If you place the min and max value of a certain parameter equal to eachother, the solver will take that parameter fixed at that value and only vary the other parameter(s).

from scipy.optimize import minimize

x0=(0.203, 5.653, 0.189089089089089) # your initial guess
bounds=((l1min,l1max),(l2min,l2max),(dmin,dmax)) # the minimum and maximum values for each parameter

res=minimize(getSpin,x0,method=‘SLSQP’,args=(),bounds=bounds,tol:1e-10) 
#res.x contains the optimized values for L1,L2 and d

answered Nov 2, 2014 at 23:09

tvandenbrande

3485 silver badges21 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

tvandenbrande Over a year ago

No problem. If the answer works, please accept it so others know it is a working solution.

Collectives™ on Stack Overflow

How to optimize the output of my python program

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related