From 48a073ce3aa069372ca11670045bb59f46dcda2c Mon Sep 17 00:00:00 2001
From: seanerice <seane.rice@gmail.com>
Date: Thu, 23 Apr 2020 12:03:33 -0400
Subject: [PATCH] Create actors.py

---
 actors.py | 142 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 142 insertions(+)
 create mode 100644 actors.py

diff --git a/actors.py b/actors.py
new file mode 100644
index 0000000..87d39ef
--- /dev/null
+++ b/actors.py
@@ -0,0 +1,142 @@
+import pandas
+from tqdm import tqdm
+
+trip_purposes = {
+    -7: "",
+    -8: "",
+    -9: "",
+    1: "Home",
+    2: "Home (paid work from home)",
+    3: "Work",
+    4: "Work trip/meeting",
+    5: "Volunteer activity",
+    6: "Drop off/pick up",
+    7: "",
+    8: "School",
+    9: "Child care",
+    10: "Adult care",
+    11: "Buy goods",
+    12: "Buy services",
+    13: "",
+    14: "",
+    15: "",
+    16: "",
+    17: "",
+    18: "",
+    19: "",
+    97: ""
+}
+
+family_income = {
+    -7: "Prefer not to answer",
+    -8: "I don't know",
+    -9: "Not ascertained",
+    1: "< $10,000",
+    2: "$10,000 to $14,999",
+    3: "$15,000 to $24,999",
+    4: "$25,000 to 34,999",
+    5: "35,000 to 49,999",
+    6: "50,000 to 74,999",
+    7: "75,000 to $99,999",
+    8: "$100,000 to $124,999",
+    9: "$125,000 to $149,999",
+    10: "$150,000 to $199,999",
+    11: "> $200,000"
+}
+
+trip_cols = ["HOUSEID", "PERSONID", "HHFAMINC", "WHYTO",
+             "WHYFROM", "STRTTIME", "ENDTIME"]
+
+
+class Trip:
+    def __init__(self, st_time, end_time, whyfrom, whyto):
+        self.start_time = st_time
+        self.end_time = end_time
+        self.whyfrom = whyfrom
+        self.whyto = whyto
+
+    def __str__(self):
+        return ", ".join([f"{k}: {v}" for k, v in self.__dict__.items()])
+
+    @staticmethod
+    def from_dfrow(dfrow):
+        st = dfrow["STRTTIME"]
+        et = dfrow["ENDTIME"]
+        src = dfrow["WHYFROM"]
+        dest = dfrow["WHYTO"]
+        trip = Trip(st, et, src, dest)
+        trip.household_id = dfrow["HOUSEID"]
+        trip.person_id = dfrow["PERSONID"]
+
+
+class SyntheticPerson:
+    def __init__(self, person_id):
+        self.id = person_id
+
+    def __str__(self):
+        s = f"Person {self.id}:\n"
+        for trip in self.trips:
+            s += "\t" + str(trip) + "\n"
+        return s
+
+    @staticmethod
+    def from_nhts_df(pid, df):
+        syn_person = SyntheticPerson(pid)
+        syn_person.trips = [Trip.from_dfrow(row) for i, row in df.iterrows()]
+        return syn_person
+
+
+class SyntheticHousehold:
+    def __init__(self, household_id):
+        self.id = household_id
+
+        self.people = []
+        self.income = 0
+
+    @staticmethod
+    def from_nhts_df(hhid, df):
+        syn_hh = SyntheticHousehold(hhid)
+        p_ids = df["PERSONID"].unique()
+        p_trips = df.groupby(["PERSONID"])
+
+        for pid, group in p_trips:
+            syn_person = SyntheticPerson.from_nhts_df(pid, group)
+            syn_hh.people.append(syn_person)
+
+        return syn_hh
+
+    def __str__(self):
+        s = f"Household {self.id}:\n"
+        for person in self.people:
+            p_str = "\t".join(("\n" + str(person)).splitlines(True))
+            s += p_str + "\n"
+        return s
+
+
+def templates(df):
+    df.sort_values(by=["HOUSEID", "PERSONID", "STRTTIME"], inplace=True)
+
+    # Filter out "other" trip purposes
+    # useable_trip_purposes = list(trip_purposes.keys())
+    # filtered = df.loc[(df["WHYFROM"].isin(useable_trip_purposes))
+    #                   & (df["WHYTO"].isin(useable_trip_purposes))]
+
+    # Split trips by household id
+    hhids = df["HOUSEID"].unique()
+    hh_trips = df.groupby(["HOUSEID"])
+
+    # Create synthetic households from dataframe
+    for hhid, group in tqdm(hh_trips, total=len(hhids)):
+        synth_hh = SyntheticHousehold.from_nhts_df(hhid, group)
+        yield synth_hh
+
+
+if __name__ == "__main__":
+    print("Reading trip data.")
+    trips_df = pandas.read_csv("data/nhts/trippub.csv", ",")
+
+    nhts_hh_templates = [hhtmp for hhtmp in templates(trips_df)]
+    # person_cols = ["HOUSEID", "PERSONID", "CNTTDTR"]
+
+    # print("Reading person data.")
+    # persons = pandas.read_csv("srtd/nhts/perpub.csv", ",")
-- 
2.51.2