Skip to content

Commit

Permalink
updates to acceptance criteria scripts
Browse files Browse the repository at this point in the history
  • Loading branch information
@AshishKuls
AshishKuls committed Jan 26, 2024
1 parent b4ffbfe commit 4249850
Show file tree
Hide file tree
Showing 4 changed files with 207 additions and 148 deletions.
67 changes: 16 additions & 51 deletions tm2py/acceptance/acceptance.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -165,6 +165,7 @@ def _make_roadway_network_comparisons(self):
s_bridge_df = self.s.simulated_bridge_details_df.copy()

o_df["time_period"] = o_df.time_period.str.lower()
#o_df = o_df.drop(columns = ["standard_link_id"])
s_trim_df = s_df[
s_df["ft"] <= self.MAX_FACILITY_TYPE_FOR_ROADWAY_COMPARISONS
].copy()
Expand Down Expand Up @@ -315,6 +316,8 @@ def _make_transit_network_comparisons(self):
.reset_index()
)

sim_df = self._fix_technology_labels(sim_df, "technology")

rail_df = pd.merge(
sim_df,
obs_df,
Expand All @@ -334,6 +337,8 @@ def _make_transit_network_comparisons(self):
)
].copy()

sim_df = self._fix_technology_labels(sim_df, "technology")

non_df = pd.merge(
sim_df,
obs_df,
Expand All @@ -342,7 +347,7 @@ def _make_transit_network_comparisons(self):
right_on=["standard_line_name", "daily_line_name", "time_period"],
)

boards_df = pd.concat([rail_df, non_df])
boards_df = pd.concat([rail_df, non_df], axis = "rows",ignore_index=True)

boards_df["operator"] = np.where(
boards_df["operator"].isnull(),
Expand All @@ -357,22 +362,23 @@ def _make_transit_network_comparisons(self):

# step 3 -- create a daily shape
df = pd.DataFrame(self.s.simulated_transit_segments_gdf).copy()
am_shape_df = df[~(df["line"].str.contains("pnr_"))].reset_index().copy()
am_shape_df = df[~(df["LINE_ID"].str.contains("pnr_"))].reset_index().copy()
am_shape_df = self.c.aggregate_line_names_across_time_of_day(
am_shape_df, "line"
am_shape_df, "LINE_ID"
)
b_df = (
am_shape_df.groupby("daily_line_name")
.agg({"line": "first"})
.agg({"LINE_ID": "first"})
.reset_index()
.copy()
)
c_df = pd.DataFrame(
self.s.simulated_transit_segments_gdf[
self.s.simulated_transit_segments_gdf["line"].isin(b_df["line"])
self.s.simulated_transit_segments_gdf["LINE_ID"].isin(b_df["LINE_ID"])
].copy()
)
daily_shape_df = pd.merge(c_df, b_df, how="left", on="line")
daily_shape_df = pd.merge(c_df, b_df, how="left", on="LINE_ID")
daily_shape_df = daily_shape_df.rename(columns={"INODE":"emme_a_node_id","JNODE":"emme_b_node_id"})

# step 4 -- join the shapes to the boardings
# for daily, join boardings to shape, as I care about the boardings more than the daily shapes
Expand All @@ -393,16 +399,16 @@ def _make_transit_network_comparisons(self):
am_join_df = pd.merge(
am_shape_df,
boards_df[boards_df["time_period"] == "am"],
how="left",
how="outer",
right_on=["line_name", "daily_line_name"],
left_on=["line", "daily_line_name"],
left_on=["LINE_ID", "daily_line_name"],
)
am_join_df["model_line_id"] = am_join_df["line_name"]
am_join_df["time_period"] = np.where(
am_join_df["time_period"].isnull(), "am", am_join_df["time_period"]
)
am_join_df = self.s._get_operator_name_from_line_name(
am_join_df, "line", "temp_operator"
am_join_df, "LINE_ID", "temp_operator"
)
am_join_df["operator"] = np.where(
am_join_df["operator"].isnull()
Expand Down Expand Up @@ -531,48 +537,7 @@ def _make_home_work_flow_comparisons(self):

def _make_zero_vehicle_household_comparisons(self):

# prepare simulated data
a_df = (
pd.merge(
self.s.simulated_zero_vehicle_hhs_df,
self.s.simulated_maz_data_df[["MAZ_ORIGINAL", "MAZSEQ"]],
left_on="maz",
right_on="MAZSEQ",
how="left",
)
.drop(columns=["maz", "MAZSEQ"])
.rename(columns={"MAZ_ORIGINAL": "maz"})
)

# TODO: probably better if this is done in Simulated, to avoid using Canonical in this class
a_df = pd.merge(
a_df,
self.c.census_2010_to_maz_crosswalk_df,
how="left",
on="maz",
)

a_df["product"] = a_df["simulated_zero_vehicle_share"] * a_df["maz_share"]

b_df = (
a_df.groupby("blockgroup")
.agg({"product": "sum", "simulated_households": "sum"})
.reset_index()
.rename(columns={"product": "simulated_zero_vehicle_share"})
)
b_df["tract"] = b_df["blockgroup"].astype("str").str.slice(stop=-1)
b_df["product"] = (
b_df["simulated_zero_vehicle_share"] * b_df["simulated_households"]
)

c_df = (
b_df.groupby("tract")
.agg({"product": "sum", "simulated_households": "sum"})
.reset_index()
)
c_df["simulated_zero_vehicle_share"] = (
c_df["product"] / c_df["simulated_households"]
)
c_df = self.s.reduced_simulated_zero_vehicle_hhs_df

# prepare the observed data
join_df = self.o.census_2017_zero_vehicle_hhs_df.copy()
Expand Down
6 changes: 1 addition & 5 deletions tm2py/acceptance/canonical.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -225,11 +225,7 @@ def _read_pems_to_link_crosswalk(self) -> pd.DataFrame:

df = pd.read_csv(os.path.join(file_root, in_file))

df["pems_station_id"] = df["station"].astype(str) + "_" + df["direction"]

assert df["pems_station_id"].is_unique # validate crosswalk - correct location?

df = df[["pems_station_id", "A", "B"]]
df = df[["station_id", "A", "B"]]

self.pems_to_link_crosswalk_df = df

Expand Down
22 changes: 11 additions & 11 deletions tm2py/acceptance/observed.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ class Observed:
census_tract_centroids_gdf: gpd.GeoDataFrame

RELEVANT_PEMS_OBSERVED_YEARS_LIST = [2014, 2015, 2016]
RELEVANT_BRIDGE_TRANSACTIONS_YEARS_LIST = [2014, 2015, 2016] # ch
RELEVANT_BRIDGE_TRANSACTIONS_YEARS_LIST = [2014, 2015, 2016]
RELEVANT_PEMS_VEHICLE_CLASSES_FOR_LARGE_TRUCK = [6, 7, 8, 9, 10, 11, 12]

ohio_rmse_standards_df = pd.DataFrame(
Expand Down Expand Up @@ -325,7 +325,7 @@ def reduce_on_board_survey(self, read_file_from_disk=True):
dtype=self.reduced_transit_on_board_df.dtypes.to_dict(),
)
else:
in_df = pd.read_feather(os.path.join(file_root, in_file))
in_df = pd.read_csv(os.path.join(file_root, in_file))
temp_df = in_df[
(in_df["weekpart"].isna()) | (in_df["weekpart"] != "WEEKEND")
].copy()
Expand Down Expand Up @@ -553,11 +553,11 @@ def _make_district_to_district_transit_flows_by_technology(self):
o_df = self.reduced_transit_spatial_flow_df.copy()
o_df = o_df[o_df["time_period"] == "am"].copy()

tm1_district_dict = self.c.taz_to_district_df.set_index("taz_tm1")[
"district_tm1"
tm2_district_dict = self.c.taz_to_district_df.set_index("taz_tm2")[
"district_tm2"
].to_dict()
o_df["orig_district"] = o_df["orig_taz"].map(tm1_district_dict)
o_df["dest_district"] = o_df["dest_taz"].map(tm1_district_dict)
o_df["orig_district"] = o_df["orig_taz"].map(tm2_district_dict)
o_df["dest_district"] = o_df["dest_taz"].map(tm2_district_dict)

for prefix in self.c.transit_technology_abbreviation_dict.keys():
o_df["{}".format(prefix.lower())] = (
Expand Down Expand Up @@ -872,9 +872,8 @@ def _reduce_pems_counts(self, read_file_from_disk=True):
self.c.pems_to_link_crosswalk_df,
out_df,
how="left",
left_on="pems_station_id",
right_on="station_id",
).drop(columns=["pems_station_id"])
on="station_id"
)

out_df = self._join_tm2_node_ids(out_df)

Expand All @@ -894,15 +893,16 @@ def _reduce_pems_counts(self, read_file_from_disk=True):
.reset_index()
)
join_df = out_df[
["A", "B", "observed_flow", "station_id", "type", "vehicle_class"]
["emme_a_node_id","emme_b_node_id", "time_period", "station_id", "type", "vehicle_class"]
].copy()
return_df = pd.merge(
median_df,
join_df,
how="left",
on=["A", "B", "observed_flow", "vehicle_class"],
on=["emme_a_node_id","emme_b_node_id", "time_period", "vehicle_class"]
).reset_index(drop=True)

#return_df = return_df.rename(columns = {"model_link_id" : "standard_link_id"})
return_df = self._join_ohio_standards(return_df)
return_df = self._identify_key_arterials_and_bridges(return_df)

Expand Down
Loading

0 comments on commit 4249850

Please sign in to comment.