consolidate chunk_tags for trip destination, tour destination, trip p…

…urpose

activitysim/abm/models/location_choice.py

@@ -108,7 +108,8 @@ def _location_sample(
         estimator,
         model_settings,
         alt_dest_col_name,
-        chunk_size, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
     """
     select a sample of alternative locations.
 
@@ -155,6 +156,7 @@ def _location_sample(
         skims=skims,
         locals_d=locals_d,
         chunk_size=chunk_size,
+        chunk_tag=chunk_tag,
         trace_label=trace_label)
 
     return choices
@@ -167,7 +169,8 @@ def location_sample(
         dest_size_terms,
         estimator,
         model_settings,
-        chunk_size, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
 
     # FIXME - MEMORY HACK - only include columns actually used in spec
     chooser_columns = model_settings['SIMULATE_CHOOSER_COLUMNS']
@@ -190,7 +193,7 @@ def location_sample(
         estimator,
         model_settings,
         alt_dest_col_name,
-        chunk_size,
+        chunk_size, chunk_tag,
         trace_label)
 
     return choices
@@ -256,7 +259,8 @@ def location_presample(
         dest_size_terms,
         estimator,
         model_settings,
-        chunk_size, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
 
     trace_label = tracing.extend_trace_label(trace_label, 'presample')
 
@@ -293,7 +297,7 @@ def location_presample(
         estimator,
         model_settings,
         DEST_TAZ,
-        chunk_size,
+        chunk_size,  chunk_tag,
         trace_label)
 
     # print(f"taz_sample\n{taz_sample}")
@@ -320,7 +324,8 @@ def run_location_sample(
         dest_size_terms,
         estimator,
         model_settings,
-        chunk_size, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
     """
     select a sample of alternative locations.
 
@@ -361,7 +366,9 @@ def run_location_sample(
             dest_size_terms,
             estimator,
             model_settings,
-            chunk_size, trace_label)
+            chunk_size,
+            chunk_tag=f'{chunk_tag}.presample',
+            trace_label=trace_label)
 
     else:
 
@@ -372,7 +379,9 @@ def run_location_sample(
             dest_size_terms,
             estimator,
             model_settings,
-            chunk_size, trace_label)
+            chunk_size,
+            chunk_tag=f'{chunk_tag}.sample',
+            trace_label=trace_label)
 
     return choices
 
@@ -383,7 +392,8 @@ def run_location_logsums(
         network_los,
         location_sample_df,
         model_settings,
-        chunk_size, trace_hh_id, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
     """
     add logsum column to existing location_sample table
 
@@ -427,6 +437,7 @@ def run_location_logsums(
         logsum_settings, model_settings,
         network_los,
         chunk_size,
+        chunk_tag,
         trace_label)
 
     # "add_column series should have an index matching the table to which it is being added"
@@ -447,7 +458,8 @@ def run_location_simulate(
         want_logsums,
         estimator,
         model_settings,
-        chunk_size, trace_label):
+        chunk_size, chunk_tag,
+        trace_label):
     """
     run location model on location_sample annotated with mode_choice logsum
     to select a dest zone from sample alternatives
@@ -506,7 +518,7 @@ def run_location_simulate(
         want_logsums=want_logsums,
         skims=skims,
         locals_d=locals_d,
-        chunk_size=chunk_size,
+        chunk_size=chunk_size, chunk_tag=chunk_tag,
         trace_label=trace_label,
         trace_choice_name=model_settings['DEST_CHOICE_COLUMN_NAME'],
         estimator=estimator)
@@ -529,7 +541,8 @@ def run_location_choice(
         want_sample_table,
         estimator,
         model_settings,
-        chunk_size, trace_hh_id, trace_label
+        chunk_size, chunk_tag,
+        trace_hh_id, trace_label
         ):
     """
     Run the three-part location choice algorithm to generate a location choice for each chooser
@@ -591,7 +604,8 @@ def run_location_choice(
                 estimator,
                 model_settings,
                 chunk_size,
-                tracing.extend_trace_label(trace_label, 'sample.%s' % segment_name))
+                chunk_tag,  # run_location_sample will add appropriate suffix for sample or presample
+                trace_label=tracing.extend_trace_label(trace_label, 'sample.%s' % segment_name))
 
         # - location_logsums
         location_sample_df = \
@@ -601,9 +615,8 @@ def run_location_choice(
                 network_los,
                 location_sample_df,
                 model_settings,
-                chunk_size,
-                trace_hh_id,
-                tracing.extend_trace_label(trace_label, 'logsums.%s' % segment_name))
+                chunk_size, chunk_tag=f'{chunk_tag}.logsums',
+                trace_label=tracing.extend_trace_label(trace_label, 'logsums.%s' % segment_name))
 
         # - location_simulate
         choices_df = \
@@ -616,8 +629,8 @@ def run_location_choice(
                 want_logsums,
                 estimator,
                 model_settings,
-                chunk_size,
-                tracing.extend_trace_label(trace_label, 'simulate.%s' % segment_name))
+                chunk_size, chunk_tag=f'{chunk_tag}.simulate',
+                trace_label=tracing.extend_trace_label(trace_label, 'simulate.%s' % segment_name))
 
         if estimator:
             if trace_hh_id:
@@ -717,6 +730,8 @@ def iterate_location_choice(
     adds annotations to persons table
     """
 
+    chunk_tag = trace_label
+
     # boolean to filter out persons not needing location modeling (e.g. is_worker, is_student)
     chooser_filter_column = model_settings['CHOOSER_FILTER_COLUMN_NAME']
 
@@ -757,7 +772,7 @@ def iterate_location_choice(
             want_sample_table=want_sample_table,
             estimator=estimator,
             model_settings=model_settings,
-            chunk_size=chunk_size,
+            chunk_size=chunk_size, chunk_tag=chunk_tag,
             trace_hh_id=trace_hh_id,
             trace_label=tracing.extend_trace_label(trace_label, 'i%s' % iteration))
 

activitysim/abm/models/trip_destination.py

@@ -56,6 +56,7 @@ def _destination_sample(
         alt_dest_col_name,
         estimator,
         chunk_size,
+        chunk_tag,
         trace_label):
     """
 
@@ -103,8 +104,9 @@ def _destination_sample(
         spec=spec,
         skims=skims,
         locals_d=locals_dict,
-        chunk_size=chunk_size,
-        trace_label=trace_label)
+        chunk_size=chunk_size, chunk_tag=chunk_tag,
+        trace_label=trace_label
+        )
 
     return choices
 
@@ -120,6 +122,8 @@ def destination_sample(
         chunk_size,
         trace_label):
 
+    chunk_tag = 'trip_destination.sample'
+
     skims = skim_hotel.sample_skims(presample=False)
     alt_dest_col_name = model_settings['ALT_DEST_COL_NAME']
 
@@ -133,7 +137,8 @@ def destination_sample(
         alt_dest_col_name,
         estimator,
         chunk_size,
-        trace_label)
+        chunk_tag=chunk_tag,
+        trace_label=trace_label)
 
     return choices
 
@@ -356,8 +361,7 @@ def destination_presample(
         trace_label):
 
     trace_label = tracing.extend_trace_label(trace_label, 'presample')
-
-    logger.info(f"{trace_label} destination_presample")
+    chunk_tag = 'trip_destination.presample'  # distinguish from trip_destination.sample
 
     alt_dest_col_name = model_settings['ALT_DEST_COL_NAME']
     maz_taz = network_los.maz_taz_df[['MAZ', 'TAZ']].set_index('MAZ').TAZ
@@ -387,7 +391,8 @@ def destination_presample(
         alt_dest_col_name,
         estimator,
         chunk_size,
-        trace_label)
+        chunk_tag=chunk_tag,
+        trace_label=trace_label)
 
     # choose a MAZ for each DEST_TAZ choice, choice probability based on MAZ size_term fraction of TAZ total
     maz_sample = choose_MAZ_for_TAZ(taz_sample, size_term_matrix, trips, network_los, alt_dest_col_name, trace_label)
@@ -423,7 +428,7 @@ def trip_destination_sample(
         pick_count : int
             number of duplicate picks for chooser, alt
     """
-    trace_label = tracing.extend_trace_label(trace_label, 'trip_destination_sample')
+    trace_label = tracing.extend_trace_label(trace_label, 'sample')
 
     assert len(trips) > 0
     assert len(alternatives) > 0
@@ -474,7 +479,8 @@ def compute_ood_logsums(
         od_skims,
         locals_dict,
         chunk_size,
-        trace_label):
+        trace_label,
+        chunk_tag):
     """
     Compute one (of two) out-of-direction logsums for destination alternatives
 
@@ -495,7 +501,8 @@ def compute_ood_logsums(
         skims=od_skims,
         locals_d=locals_dict,
         chunk_size=chunk_size,
-        trace_label=trace_label)
+        trace_label=trace_label,
+        chunk_tag=chunk_tag)
 
     assert logsums.index.equals(choosers.index)
 
@@ -526,6 +533,9 @@ def compute_logsums(
     trace_label = tracing.extend_trace_label(trace_label, 'compute_logsums')
     logger.info("Running %s with %d samples", trace_label, destination_sample.shape[0])
 
+    # chunk usage is uniform so better to combine
+    chunk_tag = 'trip_destination.compute_logsums'
+
     # FIXME should pass this in?
     network_los = inject.get_injectable('network_los')
 
@@ -588,7 +598,8 @@ def compute_logsums(
         od_skims,
         locals_dict,
         chunk_size,
-        trace_label=tracing.extend_trace_label(trace_label, 'od'))
+        trace_label=tracing.extend_trace_label(trace_label, 'od'),
+        chunk_tag=chunk_tag)
 
     # - dp_logsums
     dp_skims = {
@@ -611,7 +622,8 @@ def compute_logsums(
         dp_skims,
         locals_dict,
         chunk_size,
-        trace_label=tracing.extend_trace_label(trace_label, 'dp'))
+        trace_label=tracing.extend_trace_label(trace_label, 'dp'),
+        chunk_tag=chunk_tag)
 
     return destination_sample
 
@@ -637,6 +649,7 @@ def trip_destination_simulate(
         destination alt chosen
     """
     trace_label = tracing.extend_trace_label(trace_label, 'trip_dest_simulate')
+    chunk_tag = 'trip_destination.simulate'
 
     spec = simulate.spec_for_segment(model_settings, spec_id='DESTINATION_SPEC',
                                      segment_name=primary_purpose, estimator=estimator)
@@ -665,7 +678,7 @@ def trip_destination_simulate(
         allow_zero_probs=True, zero_prob_choice_val=NO_DESTINATION,
         skims=skims,
         locals_d=locals_dict,
-        chunk_size=chunk_size,
+        chunk_size=chunk_size, chunk_tag=chunk_tag,
         trace_label=trace_label,
         trace_choice_name='trip_dest',
         estimator=estimator)

activitysim/abm/models/trip_purpose.py

@@ -132,6 +132,9 @@ def run_trip_purpose(
     purpose: pandas.Series of purpose (str) indexed by trip_id
     """
 
+    # uniform across trip_purpose
+    chunk_tag = 'trip_purpose'
+
     model_settings_file_name = 'trip_purpose.yaml'
     model_settings = config.read_model_settings(model_settings_file_name)
 
@@ -176,7 +179,7 @@ def run_trip_purpose(
             trace_label=trace_label)
 
     for i, trips_chunk, chunk_trace_label in \
-            chunk.adaptive_chunked_choosers(trips_df, chunk_size, trace_label):
+            chunk.adaptive_chunked_choosers(trips_df, chunk_size, chunk_tag, trace_label):
 
         choices = choose_intermediate_trip_purpose(
             trips_chunk,

activitysim/abm/models/util/logsums.py

@@ -35,7 +35,9 @@ def compute_logsums(choosers,
                     tour_purpose,
                     logsum_settings, model_settings,
                     network_los,
-                    chunk_size, trace_label):
+                    chunk_size,
+                    chunk_tag,
+                    trace_label):
     """
 
     Parameters
@@ -150,6 +152,7 @@ def compute_logsums(choosers,
         skims=skims,
         locals_d=locals_dict,
         chunk_size=chunk_size,
+        chunk_tag=chunk_tag,
         trace_label=trace_label)
 
     return logsums