select_from_model_repeated

select_from_model.py

from sklearn.feature_selection import *
feat_list = [] 
all_scores = [] 
for i in range(10):
  np.random.seed(i)
  sfm = SelectFromModel(estimator = clf, threshold=None, prefit=False, 
                       norm_order=1, max_features = 12)
  sfm.fit(D[allfeats], y)
  modfeats = sfm.get_support()
  
  #print(modfeats)
  Xred = D[allfeats].iloc[:,modfeats]
  cols = list(Xred.columns)
  feat_list.append(cols)
  
  rskf = RepeatedStratifiedKFold(n_splits = 10, n_repeats = 100)
  this_score = np.mean(cross_val_score(clf, Xred, y, cv = rskf, scoring = "roc_auc"))
  print(this_score) 
  all_scores.append(this_score)

for i in range(10):
  if all_scores[i] == np.max(all_scores):
    print("These are the Best Features", feat_list[i])
    print("They scored: ", all_scores[i])
#print("Overall Average", np.mean(all_scores))