diff --git a/TASK_1/incidents_understanding_preparation.py b/TASK_1/incidents_understanding_preparation.py
index c2908ec..35373ce 100644
--- a/TASK_1/incidents_understanding_preparation.py
+++ b/TASK_1/incidents_understanding_preparation.py
@@ -168,7 +168,7 @@
 
 # %%
 print(f"# of rows before dropping duplicates: {incidents_df.shape[0]}")
-incidents_df.drop_duplicates(inplace=True) #, ignore_index=True) # TODO: geopy assume qui non sia stato resettato??
+incidents_df.drop_duplicates(inplace=True)
 print(f"# of rows after dropping duplicates: {incidents_df.shape[0]}")
 
 # %% [markdown]
@@ -518,7 +518,7 @@ def group_by_day(df, date_col):
 # To fix these inconsistencies we used the library [GeoPy]((https://geopy.readthedocs.io/en/stable/)). This library allows to retrieve the address (state, county, suburb, city, town, village, location name, and other features) corresponding to a given latitude and longitude. We queried the library using all the latitudes and longitudes of the points in the dataset and we saved the results in the CSV file we now load:
 
 # %%
-geopy_path = os.path.join(DATA_FOLDER_PATH, 'external_data/geopy.csv') # TODO: questo potraà diventare geopy (cancellare il vecchio geopy)
+geopy_path = os.path.join(DATA_FOLDER_PATH, 'external_data/geopy.csv')
 geopy_df = pd.read_csv(geopy_path, index_col=['index'], low_memory=False, dtype={})
 geopy_df.head(n=2)
 
@@ -571,7 +571,7 @@ def group_by_day(df, date_col):
     incidents_df = load_checkpoint('checkpoint_1', date_cols=['date', 'date_original'])
 else:
     geo_df = incidents_df[['state', 'city_or_county', 'address', 'latitude', 'longitude']]
-    geo_df = pd.concat([geo_df, geopy_df.loc[incidents_df.index]], axis=1) # TODO: geopy ha più righe perchè tiene anche quelle dei vecchi duplicati????
+    geo_df = pd.concat([geo_df, geopy_df.loc[incidents_df.index]], axis=1)
     geo_df = geo_df.apply(lambda row: check_geographical_data_consistency(row, additional_data=counties_df), axis=1)
     incidents_df[geo_df.columns] = geo_df[geo_df.columns]
     save_checkpoint(incidents_df, 'checkpoint_1')
@@ -1243,7 +1243,7 @@ def geodesic_distance(point1, point2):
 # We plot the boundaries of the classifier:
 
 # %%
-alabama_color_map = { #TODO: NON SO DOVE SI ROMPE, adesso in y_train ha più di 7 valori
+alabama_color_map = {
     1:'red',
     2:'orange',
     3:'yellow',
@@ -1333,7 +1333,7 @@ def geodesic_distance(point1, point2):
 # %% [markdown]
 # #### Features
 # %%
-incidents_df.groupby(['address']).size().sort_values(ascending=False)[:50].plot( #TODO: TOGLIERE(?)
+incidents_df.groupby(['address']).size().sort_values(ascending=False)[:50].plot(
     kind='bar',
     figsize=(10,6),
     title='Counts of the addresses with the 50 highest number of incidents'
@@ -1687,7 +1687,7 @@ def max_min_value(attribute): # FIXME: convertire in float, escludere <= 122 e >
 
 if LOAD_DATA_FROM_CHECKPOINT:
     with zipfile.ZipFile('checkpoints/checkpoint_4.csv.zip', 'r') as zip_ref:
-        zip_ref.extractall('checkpoints/') # TODO: magari fare all'inizio una chiamata che decomprime tutti i *.zip
+        zip_ref.extractall('checkpoints/')
     incidents_df = load_checkpoint('checkpoint_4', date_cols=['date', 'date_original'])
 else:
     new_age_df = age_temporary_df.apply(lambda row: set_gender_age_consistent_data(row), axis=1)
@@ -1957,31 +1957,6 @@ def max_min_value(attribute): # FIXME: convertire in float, escludere <= 122 e >
 # %%
 from TASK_1.data_preparation_utils import set_tags_consistent_data
 
-CASTING = {'latitude':'Float64',
-           'longitude':'Float64',
-           'participant_age1':'Int64',
-           'min_age_participants':'Int64',
-           'avg_age_participants':'Int64',
-           'max_age_participants':'Int64',
-           'n_participants_child':'Int64',
-           'n_participants_teen':'Int64',
-           'n_participants_adult':'Int64',
-           'n_males':'Int64',
-           'n_females':'Int64',
-           'n_killed':'Int64',
-           'n_injured':'Int64',
-           'n_arrested':'Int64',
-           'n_unharmed':'Int64',
-           'n_participants':'Int64',
-           'year':'Int64',
-           'month':'Int64',
-           'day':'Int64',
-           'day_of_week':'Int64',
-           'location_importance':'Float64',
-           'state_house_district':'Int64',
-           'state_senate_district':'Int64',
-           'congressional_district':'Int64'
-           }
 if LOAD_DATA_FROM_CHECKPOINT:
     with zipfile.ZipFile('checkpoints/checkpoint_6.csv.zip', 'r') as zip_ref:
         zip_ref.extractall('checkpoints/')
@@ -2315,10 +2290,9 @@ def max_min_value(attribute): # FIXME: convertire in float, escludere <= 122 e >
     'firearm', 'air_gun', 'shots', 'aggression', 'suicide', 'injuries',
     'death', 'road', 'illegal_holding', 'house', 'school', 'children',
     'drugs', 'officers', 'organized', 'social_reasons', 'defensive',
-    'workplace', 'abduction', 'unintentional'] #TODO: add tag_consistency
+    'workplace', 'abduction', 'unintentional']
 
 external_columns = ['povertyPercentage', 'party', 'candidatevotes', 'totalvotes', 'candidateperc', 'population_state_2010']
-#TODO:  majority state party?
 
 # %% [markdown]
 # We re-order the columns and we save the cleaned dataset:
@@ -2344,4 +2318,4 @@ def max_min_value(attribute): # FIXME: convertire in float, escludere <= 122 e >
 sns.heatmap(corr_matrix, mask=np.triu(corr_matrix))
 
 # %%
-incidents_df.to_csv(DATA_FOLDER_PATH +'incidents_cleaned.csv', index=False)
+incidents_df.to_csv(DATA_FOLDER_PATH +'incidents_cleaned.csv', index=True)