python scipy.optimize.minimize "SLSQP solver" adding constraint between xo

Question

I would like to optimize function with SLSQP solver. The function contains 5 parameters, and I would like to add a constraint that x[0] > x[3]. The following code makes x[0]=x[3]. Can you help me to modify it to "x[0] > x[3]"

cons = {'type':'eq', 'fun': lambda x: x[0] - x[3]} 

res = minimize(model_calib, xo, bounds=[(100,8000),(0,650),(0,1),(5,550),(0,3)], method='SLSQP',constraints = cons)

Best regards,

You should use `'type': 'ineq'` for inequality type of constraints — Cory Kramer, Jan 19 '21 at 19:00
Essentially a duplicate of: https://stackoverflow.com/questions/42303470/scipy-optimize-inequality-constraint-which-side-of-the-inequality-is-considere — Igor Rivin, Jan 19 '21 at 19:01

score 0 · Answer 1 · answered Jan 19 '21 at 19:15

0

As you mentioned in comment by @Cory Kramer

cons = {'type':'ineq', 'fun': lambda x: x[0] - x[3]}

answered Jan 19 '21 at 19:15

balezz

788
5
15

Technically speaking, that is `>= 0` not just `> 0` as the OP requested – Cory Kramer Jan 19 '21 at 20:32
May be 'lambda x: x[0] - x[3]*(1 + 1e-15)' helps here. – balezz Jan 19 '21 at 21:10
Thanks for your support @balezz May this solution give answer with x[3] > x[0]???? – Mohamed Jan 20 '21 at 06:37
I have not `model_calib` function, so can`t check this solution. Try it yourself and write what happens. – balezz Jan 20 '21 at 06:42

python scipy.optimize.minimize "SLSQP solver" adding constraint between xo

1 Answers1