add column that specifies days of the week

synthesis/population/matched.py

@@ -18,6 +18,7 @@ def configure(context):
     context.config("matching_minimum_observations", 20)
     context.config("weekend_scenario", False)
     context.config("specific_weekend_scenario", "all") # options are "all", "saturday", "sunday"
+    context.config("specific_day_scenario", "avgworkday") #options can be any of the days of the week or "avgworkday"
 
     context.stage("data.microcensus.persons")
     context.stage("synthesis.population.sampled")
@@ -161,20 +162,29 @@ def execute(context):
     df_mz = context.stage("data.microcensus.persons")
     is_weekend_scenario = context.config("weekend_scenario")
     specific_weekend_scenario = context.config("specific_weekend_scenario")
+    specific_day = context.config("specific_day_scenario")
+    is_specific_day_scenario = specific_day != "avgworkday"
 
     # Source are the MZ observations, for each STATPOP person, a sample is drawn from there
+
+    #specify day of the week the scenario will be generated for
     df_source = pd.DataFrame(df_mz[
-                                 (is_weekend_scenario & df_mz[
-                                     "weekend"])  # use only weekend samples for a weekend scenario
-                                 |
-                                 (~is_weekend_scenario & ~df_mz["weekend"])  # and only weekday samples for a weekday
-                                 ])
+                 (is_weekend_scenario & df_mz[
+                     "weekend"])  # use only weekend samples for a weekend scenario - maybe not needed
+                 |
+                 (~is_weekend_scenario & ~is_specific_day_scenario & (~df_mz["weekend"]))  # and only weekday samples for a weekday
+                 |
+                #add options for different days of the week scenarios that are not weekend
+                 (~is_weekend_scenario & (specific_day != "avgworkday") & (df_mz["day"] == specific_day))
+                 ])
 
     #If specific weekend context is needed for saturday or sunday
     if (is_weekend_scenario & (specific_weekend_scenario != "all")):
         df_source = pd.DataFrame(df_source[((specific_weekend_scenario == "saturday") & df_source["saturday"]) |
                                  ((specific_weekend_scenario == "sunday") & df_source["sunday"])])
 
+    print("INFO: The scenario will be generated for: ", df_source["day"].unique(), " day (s) of the week")
+
     df_population = context.stage("synthesis.population.sampled")
     number_of_statpop_persons = len(np.unique(df_population["person_id"]))
     number_of_statpop_households = len(np.unique(df_population["household_id"]))