fix:update

processors/days_on_market.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[2]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "days_on_market/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[3]:
+
+
+data_frames = []
+
+exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+]
+
+slug_column_mappings = {
+ "_mean_listings_price_cut_amt_": "Mean Listings Price Cut Amount",
+ "_med_doz_pending_": "Median Days on Pending",
+ "_median_days_to_pending_": "Median Days to Close",
+ "_perc_listings_price_cut_": "Percent Listings Price Cut",
+}
+
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ # skip month files for now since they are redundant
+ if "month" in filename:
+ continue
+
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+
+ if "_uc_sfrcondo_" in filename:
+ cur_df["Home Type"] = "all homes (SFR + Condo)"
+ # change column type to string
+ cur_df["RegionName"] = cur_df["RegionName"].astype(str)
+ elif "_uc_sfr_" in filename:
+ cur_df["Home Type"] = "SFR"
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[9]:
+
+
+# Adjust column names
+final_df = combined_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateName": "State",
+ }
+)
+
+final_df
+
+
+# In[5]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/for_sale_listings.py

@@ -0,0 +1,106 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[2]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "for_sale_listings/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[3]:
+
+
+exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+]
+
+slug_column_mappings = {
+ "_mlp_": "Median Listing Price",
+ "_new_listings_": "New Listings",
+ "new_pending": "New Pending",
+}
+
+
+data_frames = []
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+
+ # ignore monthly data for now since it is redundant
+ if "month" in filename:
+ continue
+
+ if "sfrcondo" in filename:
+ cur_df["Home Type"] = "all homes"
+ elif "sfr" in filename:
+ cur_df["Home Type"] = "SFR"
+ elif "condo" in filename:
+ cur_df["Home Type"] = "condo/co-op only"
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[4]:
+
+
+# Adjust column names
+final_df = combined_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateName": "State",
+ }
+)
+
+final_df
+
+
+# In[5]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/helpers.py

@@ -12,6 +12,7 @@ def coalesce_columns(
  if column_to_coalesce in col and "_" in col:
  if not pd.isna(row[col]):
  df.at[index, column_to_coalesce] = row[col]
+ continue
 
  # remove columns with underscores
  combined_df = df[columns_to_coalesce]
@@ -62,6 +63,7 @@ def get_melted_df(
  var_name="Date",
  value_name=col_name,
  )
+
  return df
 
 
@@ -70,7 +72,7 @@ def save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df):
  os.makedirs(FULL_PROCESSED_DIR_PATH)
 
  final_df.to_json(
- FULL_PROCESSED_DIR_PATH + "final.jsonl", orient="records", lines=True
+ FULL_PROCESSED_DIR_PATH + "final2.jsonl", orient="records", lines=True
  )
 
 

processors/home_value_forecasts.py

@@ -0,0 +1,96 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import get_combined_df, save_final_df_as_jsonl
+
+
+# In[2]:
+
+
+DATA_DIR = "../data/"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "home_values_forecasts/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[3]:
+
+
+data_frames = []
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+
+ cols = ["Month Over Month %", "Quarter Over Quarter %", "Year Over Year %"]
+ if filename.endswith("sm_sa_month.csv"):
+ # print('Smoothed')
+ cur_df.columns = list(cur_df.columns[:-3]) + [
+ x + " (Smoothed) (Seaonally Adjusted)" for x in cols
+ ]
+ else:
+ # print('Raw')
+ cur_df.columns = list(cur_df.columns[:-3]) + cols
+ cur_df["RegionName"] = cur_df["RegionName"].astype(str)
+
+ data_frames.append(cur_df)
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "RegionType",
+ "SizeRank",
+ "StateName",
+ "BaseDate",
+ ],
+)
+
+combined_df
+
+
+# In[4]:
+
+
+# Adjust columns
+final_df = combined_df
+final_df = combined_df.drop("StateName", axis=1)
+final_df = final_df.rename(
+ columns={
+ "CountyName": "County",
+ "BaseDate": "Date",
+ "RegionName": "Region",
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ }
+)
+
+# iterate over rows of final_df and populate State and City columns if the regionType is msa
+for index, row in final_df.iterrows():
+ if row["RegionType"] == "msa":
+ regionName = row["Region"]
+ # final_df.at[index, 'Metro'] = regionName
+
+ city = regionName.split(", ")[0]
+ final_df.at[index, "City"] = city
+
+ state = regionName.split(", ")[1]
+ final_df.at[index, "State"] = state
+
+final_df
+
+
+# In[9]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/home_values.py

@@ -0,0 +1,181 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[2]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "home_values/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[5]:
+
+
+data_frames = []
+
+slug_column_mappings = {
+ "_tier_0.0_0.33_": "Bottom Tier ZHVI",
+ "_tier_0.33_0.67_": "Mid Tier ZHVI",
+ "_tier_0.67_1.0_": "Top Tier ZHVI",
+ "": "ZHVI",
+}
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+ exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Bedroom Count",
+ "Home Type",
+ ]
+
+ if "Zip" in filename:
+ continue
+ if "Neighborhood" in filename:
+ continue
+ if "City" in filename:
+ continue
+ if "Metro" in filename:
+ continue
+ if "County" in filename:
+ continue
+
+ if "City" in filename:
+ exclude_columns = exclude_columns + ["State", "Metro", "CountyName"]
+ elif "Zip" in filename:
+ exclude_columns = exclude_columns + [
+ "State",
+ "City",
+ "Metro",
+ "CountyName",
+ ]
+ elif "County" in filename:
+ exclude_columns = exclude_columns + [
+ "State",
+ "Metro",
+ "StateCodeFIPS",
+ "MunicipalCodeFIPS",
+ ]
+ elif "Neighborhood" in filename:
+ exclude_columns = exclude_columns + [
+ "State",
+ "City",
+ "Metro",
+ "CountyName",
+ ]
+
+ if "_bdrmcnt_1_" in filename:
+ cur_df["Bedroom Count"] = "1-Bedroom"
+ elif "_bdrmcnt_2_" in filename:
+ cur_df["Bedroom Count"] = "2-Bedrooms"
+ elif "_bdrmcnt_3_" in filename:
+ cur_df["Bedroom Count"] = "3-Bedrooms"
+ elif "_bdrmcnt_4_" in filename:
+ cur_df["Bedroom Count"] = "4-Bedrooms"
+ elif "_bdrmcnt_5_" in filename:
+ cur_df["Bedroom Count"] = "5+-Bedrooms"
+ else:
+ cur_df["Bedroom Count"] = "All Bedrooms"
+
+ if "_uc_sfr_" in filename:
+ cur_df["Home Type"] = "SFR"
+ elif "_uc_sfrcondo_" in filename:
+ cur_df["Home Type"] = "all homes (SFR/condo)"
+ elif "_uc_condo_" in filename:
+ cur_df["Home Type"] = "condo"
+
+ cur_df["StateName"] = cur_df["StateName"].astype(str)
+ cur_df["RegionName"] = cur_df["RegionName"].astype(str)
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Bedroom Count",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[11]:
+
+
+final_df = combined_df
+
+for index, row in final_df.iterrows():
+ if row["RegionType"] == "city":
+ final_df.at[index, "City"] = row["RegionName"]
+ elif row["RegionType"] == "county":
+ final_df.at[index, "County"] = row["RegionName"]
+ if row["RegionType"] == "state":
+ final_df.at[index, "StateName"] = row["RegionName"]
+
+# coalesce State and StateName columns
+# final_df["State"] = final_df["State"].combine_first(final_df["StateName"])
+# final_df["County"] = final_df["County"].combine_first(final_df["CountyName"])
+
+# final_df = final_df.drop(
+# columns=[
+# "StateName",
+# # "CountyName"
+# ]
+# )
+final_df
+
+
+# In[12]:
+
+
+final_df = final_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateCodeFIPS": "State Code FIPS",
+ "StateName": "State",
+ "MunicipalCodeFIPS": "Municipal Code FIPS",
+ }
+)
+
+final_df
+
+
+# In[13]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/new_construction.py

@@ -0,0 +1,101 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[2]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "new_construction/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[3]:
+
+
+exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+]
+
+slug_column_mappings = {
+ "_median_sale_price_per_sqft": "Median Sale Price per Sqft",
+ "_median_sale_price": "Median Sale Price",
+ "sales_count": "Sales Count",
+}
+
+data_frames = []
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+
+ if "sfrcondo" in filename:
+ cur_df["Home Type"] = "all homes"
+ elif "sfr" in filename:
+ cur_df["Home Type"] = "SFR"
+ elif "condo" in filename:
+ cur_df["Home Type"] = "condo/co-op only"
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[4]:
+
+
+final_df = combined_df
+final_df = final_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateName": "State",
+ }
+)
+
+final_df.sort_values(by=["Region ID", "Home Type", "Date"])
+
+
+# In[5]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/rentals.py

@@ -0,0 +1,160 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[2]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[3]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "rentals/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[7]:
+
+
+data_frames = []
+
+slug_column_mappings = {"": "Rent"}
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ # print("processing " + filename)
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+ exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ ]
+
+ if "_sfrcondomfr_" in filename:
+ cur_df["Home Type"] = "all homes plus multifamily"
+ # change column type to string
+ cur_df["RegionName"] = cur_df["RegionName"].astype(str)
+ if "City" in filename:
+ exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ # City Specific
+ "State",
+ "Metro",
+ "CountyName",
+ ]
+ elif "Zip" in filename:
+ exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ # Zip Specific
+ "State",
+ "City",
+ "Metro",
+ "CountyName",
+ ]
+ elif "County" in filename:
+ exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ # County Specific
+ "State",
+ "Metro",
+ "StateCodeFIPS",
+ "MunicipalCodeFIPS",
+ ]
+
+ elif "_sfr_" in filename:
+ cur_df["Home Type"] = "SFR"
+ elif "_mfr_" in filename:
+ cur_df["Home Type"] = "multifamily"
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[8]:
+
+
+final_df = combined_df
+
+for index, row in final_df.iterrows():
+ if row["RegionType"] == "city":
+ final_df.at[index, "City"] = row["RegionName"]
+ elif row["RegionType"] == "county":
+ final_df.at[index, "County"] = row["RegionName"]
+
+# coalesce State and StateName columns
+final_df["State"] = final_df["State"].combine_first(final_df["StateName"])
+final_df["State"] = final_df["County"].combine_first(final_df["CountyName"])
+
+final_df = final_df.drop(columns=["StateName", "CountyName"])
+final_df
+
+
+# In[6]:
+
+
+# Adjust column names
+final_df = final_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateCodeFIPS": "State Code FIPS",
+ "MunicipalCodeFIPS": "Municipal Code FIPS",
+ }
+)
+
+final_df
+
+
+# In[7]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

processors/sales.py

@@ -0,0 +1,106 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import os
+
+from helpers import (
+ get_combined_df,
+ save_final_df_as_jsonl,
+ handle_slug_column_mappings,
+)
+
+
+# In[2]:
+
+
+DATA_DIR = "../data"
+PROCESSED_DIR = "../processed/"
+FACET_DIR = "sales/"
+FULL_DATA_DIR_PATH = os.path.join(DATA_DIR, FACET_DIR)
+FULL_PROCESSED_DIR_PATH = os.path.join(PROCESSED_DIR, FACET_DIR)
+
+
+# In[3]:
+
+
+exclude_columns = [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+]
+
+slug_column_mappings = {
+ "_median_sale_to_list_": "Median Sale to List Ratio",
+ "_mean_sale_to_list_": "Mean Sale to List Ratio",
+ "_median_sale_price_": "Median Sale Price",
+ "_pct_sold_above_list_": "% Sold Above List",
+ "_pct_sold_below_list_": "% Sold Below List",
+ "_sales_count_now_": "Nowcast",
+}
+
+data_frames = []
+
+for filename in os.listdir(FULL_DATA_DIR_PATH):
+ if filename.endswith(".csv"):
+ print("processing " + filename)
+ # ignore monthly data for now since it is redundant
+ if "month" in filename:
+ continue
+
+ cur_df = pd.read_csv(os.path.join(FULL_DATA_DIR_PATH, filename))
+
+ if "_sfrcondo_" in filename:
+ cur_df["Home Type"] = "all homes"
+ elif "_sfr_" in filename:
+ cur_df["Home Type"] = "SFR"
+
+ data_frames = handle_slug_column_mappings(
+ data_frames, slug_column_mappings, exclude_columns, filename, cur_df
+ )
+
+
+combined_df = get_combined_df(
+ data_frames,
+ [
+ "RegionID",
+ "SizeRank",
+ "RegionName",
+ "RegionType",
+ "StateName",
+ "Home Type",
+ "Date",
+ ],
+)
+
+combined_df
+
+
+# In[4]:
+
+
+# Adjust column names
+final_df = combined_df.rename(
+ columns={
+ "RegionID": "Region ID",
+ "SizeRank": "Size Rank",
+ "RegionName": "Region",
+ "RegionType": "Region Type",
+ "StateName": "State",
+ }
+)
+
+final_df.sort_values(by=["Region ID", "Home Type", "Date"])
+
+
+# In[5]:
+
+
+save_final_df_as_jsonl(FULL_PROCESSED_DIR_PATH, final_df)
+

zillow.py

@@ -299,7 +299,7 @@ class Zillow(datasets.GeneratorBasedBuilder):
  )
 
  def _split_generators(self, dl_manager):
- file_path = os.path.join("processed", self.config.name, "final1.jsonl")
+ file_path = os.path.join("processed", self.config.name, "final2.jsonl")
  file_train = dl_manager.download(file_path)
  # file_test = dl_manager.download(os.path.join(self.config.name, "test.csv"))
  # file_eval = dl_manager.download(os.path.join(self.config.name, "valid.csv"))