Scipy optimize minimize performs only 1 iteration

Question

I 'am trying to optimize a function which takes an input list of weights and change pixels of images with those weights and then calculates some Score (NSS Score) . Our target is to find the best input weights that give the highest NSS Score . For that we are using the Scipy minimize function

    # Our input initial Weights , which we want to find the highest values of them
    weights_init= [i for i in range(82)]
    weights_init[0] = 0 

    min_result= minimize(global_stimuli_optimization,weights_init,method='L-BFGS-B', options={'disp': True})

The global_stimuli_optimization function takes the weights als input and make some processing and then calculate some Score (NSS)

def global_stimuli_optimization(weights):
    
    ... some attributs ... 

    for index,boundary in enumerate (boundaries):
        ... Some processing  ...
                
    
    all_stimuli = [all_stimuli_sorted for _,all_stimuli_sorted in sorted(zip(indices_sorted,all_stimuli_sorted))]

    # we create a model
    model  = MySaliencyMapModel(all_stimuli)
    # we calculate a score
    nss_score = np.nanmean(model.NSSs(mit_stimuli, mit_fixations))
    
    print('Weights : ',weights,' NSS :' , nss_score)
    
    return  - nss_score

The optimize functions works fine BUT performs only 1 iteration and then stops

Output :


           * * *

Tit   = total number of iterations
Tnf   = total number of function evaluations
Tnint = total number of segments explored during Cauchy searches
Skip  = number of BFGS updates skippeds
Nact  = number of active bounds at final generalized Cauchy point
Projg = norm of the final projected gradient
F     = final function value

           * * *

   N    Tit     Tnf  Tnint  Skip  Nact     Projg        F
   82      0      1      0     0     0   0.000D+00  -6.932D-01
  F = -0.693229780993317 


CONVERGENCE: NORM_OF_PROJECTED_GRADIENT_<=_PGTOL

 Cauchy                time 0.000E+00 seconds.
 Subspace minimization time 0.000E+00 seconds.
 Line search           time 0.000E+00 seconds.

 Total User time 0.000E+00 seconds.

Does anyone have an idea which it stops after the first iteration ?

Scipy optimize minimize performs only 1 iteration

0 Answers