Ensure always subset stop id during wait time calc (#138)

* Update how calc wait time when no dir id

* lint

peartree/graph_tool.py

@@ -40,7 +40,7 @@ class GraphToolNotImported(Exception):
 def get_prop_type(value: Any, key: Any=None) -> Tuple[str, Any, str]:
     """
     Performs typing and value conversion for the graph_tool PropertyMap class.
-    
+
     If a key is provided, it also ensures the key is in a format that can be
     used with the PropertyMap. Returns a tuple, (type name, value, key)
     """
@@ -102,7 +102,7 @@ def nx_to_gt(nxG: nx.MultiDiGraph):
         The converted network graph, instantiated as a graph_tool network graph
     """
     # First, attempt to import graph-tool
-    gt = _import_graph_tool()        
+    gt = _import_graph_tool()
 
     # Phase 0: Create a directed or undirected graph-tool Graph
     gtG = gt.Graph(directed=nxG.is_directed())
@@ -112,14 +112,14 @@ def nx_to_gt(nxG: nx.MultiDiGraph):
         # Convert the value and key into a type for graph-tool
         tname, value, key = get_prop_type(value, key)
 
-        prop = gtG.new_graph_property(tname) # Create the PropertyMap
+        prop = gtG.new_graph_property(tname)  # Create the PropertyMap
         gtG.graph_properties[key] = prop     # Set the PropertyMap
         gtG.graph_properties[key] = value    # Set the actual value
 
     # Phase 1: Add the vertex and edge property maps
     # Go through all nodes and edges and add seen properties
     # Add the node properties first
-    nprops = set() # cache keys to only add properties once
+    nprops = set()  # cache keys to only add properties once
     for node, data in nxG.nodes(data=True):
         # Go through all the properties if not seen and add them.
         for key, val in data.items():
@@ -128,7 +128,7 @@ def nx_to_gt(nxG: nx.MultiDiGraph):
                 continue
 
             # Convert the value and key into a type for graph-tool
-            tname, _, key  = get_prop_type(val, key)
+            tname, _, key = get_prop_type(val, key)
 
             # Create the PropertyMap, and...
             prop = gtG.new_vertex_property(tname)
@@ -145,7 +145,7 @@ def nx_to_gt(nxG: nx.MultiDiGraph):
     gtG.vertex_properties['id'] = gtG.new_vertex_property('string')
 
     # Add the edge properties second
-    eprops = set() # cache keys to only add properties once
+    eprops = set()  # cache keys to only add properties once
     for src, dst, data in nxG.edges(data=True):
 
         # Go through all the edge properties if not seen and add them,
@@ -190,7 +190,7 @@ def nx_to_gt(nxG: nx.MultiDiGraph):
 
         # Add the edge properties
         for key, value in data.items():
-            gtG.ep[key][e] = value # ep is short for edge_properties
+            gtG.ep[key][e] = value  # ep is short for edge_properties
 
     # Done, finally!
     return gtG

peartree/parallel.py

@@ -4,7 +4,6 @@
 import numpy as np
 import pandas as pd
 from peartree.toolkit import nan_helper
-from peartree.utilities import log
 
 
 class RouteProcessorManager(BaseManager):
@@ -75,7 +74,7 @@ def generate_route_costs(self, route_id: str):
         # Check direction_id column value before using
         # trips_and_stop_times to generate wait and edge costs
         # Note: Advantage to adding handling at route level is that peartree
-        #       avoids tossing direction id if a specific route has all direction
+        #       avoids tossing direction id if specific route has all direction
         #       id rows filled in (while another does not, which is possible).
         if 'direction_id' in trips_and_stop_times:
             # If there is such column then check if it contains NaN
@@ -96,7 +95,7 @@ def generate_route_costs(self, route_id: str):
         # Look up wait time for each stop in wait_times for each direction
         wait_zero = stop_id_col.apply(lambda x: wait_times[0][x])
         trips_and_stop_times['wait_dir_0'] = wait_zero
-        
+
         wait_one = stop_id_col.apply(lambda x: wait_times[1][x])
         trips_and_stop_times['wait_dir_1'] = wait_one
 
@@ -117,26 +116,29 @@ def generate_wait_times(
         stop_cost_method: Any) -> Dict[int, List[float]]:
     wait_times = {0: {}, 1: {}}
     for stop_id in trips_and_stop_times.stop_id.unique():
+        constraint_1 = (trips_and_stop_times.stop_id == stop_id)
+        stop_times_by_stop = trips_and_stop_times[constraint_1]
+
         # Handle both inbound and outbound directions
         for direction in [0, 1]:
             # Check if direction_id exists in source data
             if 'direction_id' in trips_and_stop_times:
-                constraint_1 = (trips_and_stop_times.direction_id == direction)
-                constraint_2 = (trips_and_stop_times.stop_id == stop_id)
-                both_constraints = (constraint_1 & constraint_2)
-                direction_subset = trips_and_stop_times[both_constraints]
+                constraint_2 = (trips_and_stop_times.direction_id == direction)
+                direction_subset = stop_times_by_stop[constraint_2]
             else:
-                direction_subset = trips_and_stop_times.copy()
+                direction_subset = stop_times_by_stop.copy()
 
             if direction_subset.empty:
                 # Cannot calculate the average wait time if there are no
                 # values associated with the specified direction so default NaN
                 average_wait = np.nan
             else:
+                arrival_times_copy = np.array(direction_subset.arrival_time)
+                arrival_times_copy.sort()
                 average_wait = stop_cost_method(
                     target_time_start,
                     target_time_end,
-                    direction_subset.arrival_time)
+                    arrival_times_copy)
 
             # Add according to which direction we are working with
             wait_times[direction][stop_id] = average_wait

peartree/paths.py

@@ -102,7 +102,7 @@ def get_representative_feed(file_loc: str,
                           'were identified in GTFS.')
 
     # TODO: Due to partridge's assertion error being raised, this
-    #       check may no longer be needed.    
+    #       check may no longer be needed.
     if not len(trip_counts_by_date.items()):
         # Otherwise, error out
         raise InvalidGTFS('No valid trip counts by date '

peartree/plot.py

@@ -1,5 +1,3 @@
-import os
-
 import networkx as nx
 
 from .utilities import log

peartree/summarizer.py

@@ -11,10 +11,10 @@
 from peartree.utilities import log
 
 
-
 class InvalidParsedWaitTimes(Exception):
     pass
 
+
 def _format_summarized_outputs(summarized: pd.Series) -> pd.DataFrame:
     # The output of the group by produces a Series, but we want to extract
     # the values from the index and the Series itself and generate a

tests/test_convert.py

@@ -1,8 +1,8 @@
 import os
 
 import networkx as nx
-from peartree.paths import get_representative_feed, load_feed_as_graph
 from peartree.convert import convert_to_digraph
+from peartree.paths import get_representative_feed, load_feed_as_graph
 
 
 def fixture(filename):

tests/test_graph.py

@@ -1,11 +1,10 @@
 import os
-import pytest
 
+import pytest
 from peartree.graph import (generate_empty_md_graph,
                             generate_summary_graph_elements)
 from peartree.paths import (FALLBACK_STOP_COST_DEFAULT,
-                            _calculate_means_default,
-                            get_representative_feed)
+                            _calculate_means_default, get_representative_feed)
 
 
 def fixture(filename):

tests/test_graph_assembly.py

@@ -3,8 +3,7 @@
 
 from peartree.graph import generate_empty_md_graph, populate_graph
 from peartree.paths import (FALLBACK_STOP_COST_DEFAULT,
-                            _calculate_means_default,
-                            get_representative_feed)
+                            _calculate_means_default, get_representative_feed)
 from peartree.summarizer import (generate_edge_and_wait_values,
                                  generate_summary_edge_costs,
                                  generate_summary_wait_times)

tests/test_graph_tool.py

@@ -1,6 +1,5 @@
 import json
 import os
-import pytest
 
 from peartree.graph_tool import _import_graph_tool, nx_to_gt
 from peartree.paths import load_synthetic_network_as_graph
@@ -30,7 +29,7 @@ def test_conversion_to_graph_tool():
     # Make sure that the result is indeed a graph-tool Graph class object
     gt = _import_graph_tool()
     assert isinstance(gtG, gt.Graph)
-    
+
     # Also make sure that the attributes for all parameters have been preserved
     assert set(gtG.vp.keys()) == set(('boarding_cost', 'id', 'x', 'y'))
     assert set(gtG.gp.keys()) == set(('crs', 'name'))

tests/test_paths.py

@@ -1,5 +1,4 @@
 import json
-import math
 import os
 
 import geopandas as gpd
@@ -319,7 +318,6 @@ def test_synthetic_stop_assignment_adjustment():
     # The new graph should now be just (495 - 293) nodes
     assert len(G2.nodes()) == 495 - 293
 
-
     # We will use this as a reference distance limit for checking
     # nearest points to ensure there are some adjusted stops matches
     dist_limit = target_stop_dist_override * 0.5