diff --git a/UQ_FS_eg.py b/UQ_FS_eg.py
index a5ae568..cc717d7 100644
--- a/UQ_FS_eg.py
+++ b/UQ_FS_eg.py
@@ -60,43 +60,51 @@ fs_bmod = clf2.best_estimator_
 print('\nbest model with feature selection:', fs_bmod)
 
 #########################################################
+#cv = rskf_cv
+cv = skf_cv
+
 # my data: Feature Selelction + GridSearch CV + Pipeline
 pipe = Pipeline([
     ('pre', MinMaxScaler())
-    , ('fs', RFECV(LogisticRegression(**rs), cv = rskf_cv, scoring = 'matthews_corrcoef'))
+#    , ('fs', RFECV(LogisticRegression(**rs), cv = cv, scoring = 'matthews_corrcoef'))
+    , ('fs', RFECV(DecisionTreeClassifier(**rs), cv = cv, scoring = 'matthews_corrcoef'))
+
     , ('clf',  LogisticRegression(**rs))])
 
-search_space = [{'fs__min_features_to_select': [1,2]
-                # ,'fs__cv': [rskf_cv]
-                },
-                {
-                #'clf': [LogisticRegression()],
-                 #'clf__C': np.logspace(0, 4, 10),
-                 'clf__C': [1],
-                 'clf__max_iter': [100],
-                 'clf__penalty': ['l1', 'l2'],
-                 'clf__solver': ['saga']
-                 },
+search_space = [
+    { 'fs__estimator': [LogisticRegression(**rs)]
+     , 'fs__min_features_to_select': [0,1]
+     ,'fs__cv': [rskf_cv]
+     },
+    {
+     #'clf': [LogisticRegression()],
+     #'clf__C': np.logspace(0, 4, 10),
+     'clf__C': [1],
+     'clf__max_iter': [100],
+     'clf__penalty': ['l1', 'l2'],
+     'clf__solver': ['saga']
+     },
                 
-                {
-                #'clf': [LogisticRegression()],
-                 #'clf__C': np.logspace(0, 4, 10),
-                 'clf__C': [2, 2.5],
-                 'clf__max_iter': [100],
-                 'clf__penalty': ['l1', 'l2'],
-                 'clf__solver': ['saga']
-                 },
-                #{'clf': [RandomForestclf(n_estimators=100)],
-                # 'clf__max_depth': [5, 10, None]},
-                #{'clf': [KNeighborsclf()],
-                # 'clf__n_neighbors': [3, 7, 11],
-                # 'clf__weights': ['uniform', 'distance']
-                #}
+    {
+     #'clf': [LogisticRegression()],
+     #'clf__C': np.logspace(0, 4, 10),
+     'clf__C': [2, 2.5],
+     'clf__max_iter': [100],
+     'clf__penalty': ['l1', 'l2'],
+     'clf__solver': ['saga']
+     },
+                
+    #{'clf': [RandomForestclf(n_estimators=100)],
+    # 'clf__max_depth': [5, 10, None]},
+    #{'clf': [KNeighborsclf()],
+    # 'clf__n_neighbors': [3, 7, 11],
+    # 'clf__weights': ['uniform', 'distance']
+    #}
                 ]
 
 gscv_fs = GridSearchCV(pipe
                    , search_space
-                   , cv = rskf_cv
+                   , cv = cv
                    , scoring = mcc_score_fn
                    , refit = 'mcc'
                    , verbose = 1
@@ -111,14 +119,21 @@ gscv_fs.best_score_
 # Training best score corresponds to the max of the mean_test<score>
 train_bscore = round(gscv_fs.best_score_, 2); train_bscore
 print('\nTraining best score (MCC):', train_bscore)
+gscv_fs.cv_results_['mean_test_mcc']
 round(gscv_fs.cv_results_['mean_test_mcc'].max(),2)
+round(np.nanmax(gscv_fs.cv_results_['mean_test_mcc']),2)
+
+check_train_score = [round(gscv_fs.cv_results_['mean_test_mcc'].max(),2)
+                    , round(np.nanmax(gscv_fs.cv_results_['mean_test_mcc']),2)]
+
+check_train_score = np.nanmax(check_train_score)
 
 # Training results
 gscv_tr_resD = gscv_fs.cv_results_
 mod_refit_param =  gscv_fs.refit
 
 # sanity check
-if train_bscore == round(gscv_tr_resD['mean_test_mcc'].max(),2):
+if train_bscore == check_train_score:
     print('\nVerified training score (MCC):', train_bscore )
 else:
     print('\nTraining score could not be internatlly verified. Please check training results dict')
@@ -186,19 +201,27 @@ print('\n========================================'
 bts_predict = gscv_fs.predict(X_bts)
 print('\nMCC on Blind test:'     , round(matthews_corrcoef(y_bts, bts_predict),2))
 print('\nAccuracy on Blind test:', round(accuracy_score(y_bts, bts_predict),2))
+bts_mcc_score = round(matthews_corrcoef(y_bts, bts_predict),2)
+
+# Diff b/w train and bts test scores
+train_test_diff = train_bscore - bts_mcc
+print('\nDiff b/w train and blind test score (MCC):', train_test_diff)
+
 
 # create a dict with all scores
 lr_btsD = {#'best_model': list(gscv_lr_fit_be_mod.items())
-               'bts_fscore':None
-               , 'bts_mcc':None
+               #'bts_mcc':None
+                'bts_fscore':None
                , 'bts_precision':None
                , 'bts_recall':None
                , 'bts_accuracy':None
                , 'bts_roc_auc':None
-               , 'bts_jaccard':None }
+               , 'bts_jaccard':None}
+
+
 lr_btsD
+#lr_btsD['bts_mcc']       = bts_mcc_score
 lr_btsD['bts_fscore']    = round(f1_score(y_bts, bts_predict),2)
-lr_btsD['bts_mcc']       = round(matthews_corrcoef(y_bts, bts_predict),2)
 lr_btsD['bts_precision'] = round(precision_score(y_bts, bts_predict),2)
 lr_btsD['bts_recall']    = round(recall_score(y_bts, bts_predict),2)
 lr_btsD['bts_accuracy']  = round(accuracy_score(y_bts, bts_predict),2)
@@ -218,8 +241,7 @@ output_modelD = {'model_name': model_name
                  , 'fs_res_array_rank': gscv_fs.best_estimator_.named_steps['fs'].ranking_
                  , 'all_feature_names': all_features
                  , 'n_sel_features': n_sf
-                 , 'sel_features_names': sel_features
-                 , 'train_score (MCC)': train_bscore}
+                 , 'sel_features_names': sel_features}
 output_modelD
 
 #========================================
@@ -228,6 +250,12 @@ output_modelD
 output_modelD.update(lr_btsD)
 output_modelD
 
+output_modelD['train_score (MCC)'] = train_bscore
+output_modelD['bts_mcc'] = bts_mcc_score
+output_modelD['train_bts_diff'] = train_test_diff
+output_modelD
+
+
 #========================================
 # Write final output file
 # https://stackoverflow.com/questions/19201290/how-to-save-a-dictionary-to-a-file