diff --git a/notebooks/single_dataset_pathlib.py b/notebooks/single_dataset_pathlib.py
index 8a8b32a..7a79edd 100644
--- a/notebooks/single_dataset_pathlib.py
+++ b/notebooks/single_dataset_pathlib.py
@@ -7,8 +7,8 @@
 
 
 # Set environment variables
-# os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"
-os.environ["CUDA_VISIBLE_DEVICES"] = "1"
+os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"
+os.environ["CUDA_VISIBLE_DEVICES"] = "0"
 
 import cryovesnet
 
@@ -21,39 +21,46 @@
 
 
 pl2 = cryovesnet.Pipeline(dataset_directory,pattern='*.rec.nad')
-# pl2.setup_cryovesnet_dir(initialize=False, make_masks=True)
-# #
-# #
+pl2.setup_cryovesnet_dir(initialize=False, make_masks=True)
 #
-# # pl2.run_deep(force_run=True, gauss=True, rescale= None,  weight_path=None)
-# pl2.run_deep(force_run=True,augmentation_level=4)
-# # pl2.run_deep(force_run=True, gauss=False, rescale= None, augmentation_level=4,  weight_path=None)
-# pl2.rescale(force_run=True, slice_range=None)
-# #pl2.rescale(force_run=True,slice_range=[1,130])
-#
-# pl2.label_vesicles(input_array_name="deep_mask", within_segmentation_region=True,threshold_coef=None)
-# # pl2.label_vesicles(input_array_name="deep_mask", within_segmentation_region=False,threshold_coef=0.9)
-#
-# pl2.label_vesicles_adaptive(expanding = False, convex=False, separating=True)
-# df = pl2.make_spheres(tight=True, keep_ellipsoid=False)
-# pl2.repair_spheres()
-#
-#
-# #pl2.make_full_modfile(input_array_name='convex_labels')
 #
-# # pl2.last_output_array_name="convex_labels"
-# # pl2.last_output_array_name= "mancorr_labels"
-# # pl2.fix_spheres_interactively()
-# # pl2.fix_spheres_interactively(max_expected_diameter=45)
-# # pl2.fix_spheres_interactively("mancorr_labels")
-#
-#
-# # pl2.last_output_array_name="convex_labels"
-# os.system('rm -rf /media/amin/mtwo/84/pyto')
-# pl2.make_full_modfile(input_array_name='convex_labels')
+
+# pl2.run_deep(force_run=True, gauss=True, rescale= None,  weight_path=None)
+# pl2.run_deep(force_run=True,augmentation_level=4)
+pl2.run_deep(force_run=True,augmentation_level=1)
+# pl2.run_deep(force_run=True, gauss=False, rescale= None, augmentation_level=4,  weight_path=None)
+pl2.rescale(force_run=True, slice_range=None)
+#pl2.rescale(force_run=True,slice_range=[1,130])
+
+pl2.label_vesicles(input_array_name="deep_mask", within_segmentation_region=True,threshold_coef=None)
+# pl2.label_vesicles(input_array_name="deep_mask", within_segmentation_region=False,threshold_coef=0.9)
+
+pl2.label_vesicles_adaptive(expanding = False, convex=False, separating=True)
+df = pl2.make_spheres(tight=False, keep_ellipsoid=False)
+pl2.repair_spheres()
+
+
+#pl2.make_full_modfile(input_array_name='convex_labels')
+
+### For manual correction ###
 pl2.last_output_array_name="convex_labels"
-pl2.make_full_label_file()
-pl2.pyto_wrapper()
+pl2.fix_spheres_interactively(max_expected_diameter=45)
 
 
-#
+### In case you want to continue the manual correction ###
+# pl2.last_output_array_name= "mancorr_labels"
+# pl2.fix_spheres_interactively()
+### or:
+# pl2.fix_spheres_interactively("mancorr_labels")
+
+
+### To clean the old files ###
+#os.system('rm -rf /media/amin/mtwo/84/pyto')
+
+## In case u jsut want the automatic segmentation without manual correction
+## pl2.last_output_array_name="convex_labels"
+
+### For connector segmentation ###
+# pl2.make_full_modfile()
+# pl2.make_full_label_file()
+# pl2.pyto_wrapper()