def clean_stl_add(df):
df = df.rename(
columns={
"STREETNAME": "address_sn",
"STREETTYPE": "address_ss",
"PREDIR": "address_sd",
"ZIP_CODE": "address_zip"
})
df['index'] = np.arange(df.shape[0])
df = df.to_crs(default_crs)
df.crs = default_crs
bounds = df.bounds
df['address_city'] = 'saint louis'
df['latitude_min'] = bounds["miny"]
df['latitude_max'] = bounds["maxy"]
df['longitude_min'] = bounds["minx"]
df['longitude_max'] = bounds["maxx"]
df['direction'] = np.where(
((df['FROMLEFT'] < df['TOLEFT']) & (df['FROMRIGHT'] < df['TORIGHT'])),
"NE",
np.where(((df['FROMLEFT'] < df['TOLEFT']) &
(df['FROMRIGHT'] > df['TORIGHT'])), "NW",
np.where(((df['FROMLEFT'] > df['TOLEFT']) &
(df['FROMRIGHT'] < df['TORIGHT'])), "SE",
np.where(((df['FROMLEFT'] > df['TOLEFT']) &
(df['FROMRIGHT'] > df['TORIGHT'])), "SW",
"SW"))))
df_r = df[[col for col in df.columns if not bool(re.search("LEFT", col))]]
df_r['address_n1'] = np.where(df_r['FROMRIGHT'] > df_r['TORIGHT'],
df_r['TORIGHT'], df_r['FROMRIGHT'])
df_r['address_n2'] = np.where(df_r['TORIGHT'] > df_r['FROMRIGHT'],
df_r['TORIGHT'], df_r['FROMRIGHT'])
df_l = df[[col for col in df.columns if not bool(re.search("RIGHT", col))]]
df_l['address_n1'] = np.where(df_l['FROMLEFT'] > df_l['TOLEFT'],
df_l['TOLEFT'], df_l['FROMLEFT'])
df_l['address_n2'] = np.where(df_l['TOLEFT'] > df_l['FROMLEFT'],
df_l['TOLEFT'], df_l['FROMLEFT'])
df = pd.concat([df_r, df_l])
df = df[~((df['address_n1'] <= 0) & (df['address_n1'] <= 0))]
df = make_panel(df,
start_year="address_n1",
end_year="address_n2",
current_year=df['address_n2'],
evens_and_odds=True).rename(columns={'year': 'address_n1'})
# interpolate lat long
df = interpolate_polygon(df, "index", "direction")
df['lat'] = df['lat_interpolated']
df['long'] = df["long_interpolated"]
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_chi_add_parcels(df):
chicago_rename_dict = {
'property_address': 'address_fa',
'property_city': 'address_city',
'property_zip': 'address_zip',
'pin': 'parcelID',
'latitude': 'lat',
'longitude': 'long'
}
df.rename(columns=chicago_rename_dict, inplace=True)
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=4)
return df
def clean_sac_add(df):
sac_rename_dict = {
'APN': 'parcelID',
"Address_Number": 'address_n1',
"Street_Name": "address_sn",
"Street_Suffix": "address_ss",
"Pre_Directiona;": "address_sd",
"Postal_City": 'address_city',
"Zip_Code": "address_zip",
"Latitude_Y": "lat",
"Longitude_X": "long",
}
df.rename(columns=sac_rename_dict, inplace=True)
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_chi_add_points(df):
chicago_rename_dict = {
'ADDRNOCOM': 'address_n1',
'STNAMEPRD': 'address_sd',
'STNAME': 'address_sn',
'STNAMEPOT': 'address_ss',
'PLACENAME': 'address_city',
'ZIP5': 'address_zip',
'CMPADDABRV': 'address_fa',
'PIN': 'parcelID',
'XPOSITION': 'long',
'YPOSITION': 'lat'
}
df.rename(columns=chicago_rename_dict, inplace=True)
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_baton_rouge_add(df):
baton_rouge_rename_dict = {
'ADDRNOCOM': 'address_n1',
'ASTREET PREFIX DIRECTION': 'address_sd',
'STREET NAME': 'address_sn',
'STREET SUFFIX TYPE': 'address_ss',
'CITY': 'address_city',
'ZIP': 'address_zip',
'FULL ADDRESS': 'address_fa'
}
df.rename(columns=baton_rouge_rename_dict, inplace=True)
lat_long = df['GEOLOCATION'].str.extract('([0-9\.]+),([0-9\.]+)')
df['lat'] = lat_long.iloc[:, 0]
df['long'] = lat_long.iloc[:, 1]
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=4)
return df
def clean_seattle_add(df):
seattle_rename_dict = {
'PIN': 'parcelID',
'ADDR_NUM': 'address_n1',
'ADDR_SN': 'address_sn',
'ADDR_ST': 'address_ss',
'ADDR_SD': 'address_sd',
'ZIP5': 'address_zip',
'CTYNAME': 'address_city',
'ADDR_FULL': 'address_fa',
'LON': 'long',
'LAT': 'lat'
}
df.rename(columns=seattle_rename_dict, inplace=True)
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_orlando_add(df):
orlando_rename_dict = {
'OFFICIAL_P': 'parcelID',
"COMPLETE_A": 'address_fa',
"ADDRESS__1": 'address_n1',
"ADDRESS__2": "address_n2",
"BASENAME": "address_sn",
"POST_TYPE": "address_ss",
"POST_DIREC": "address_sd",
"MUNICIPAL_": 'address_city',
"ZIPCODE": "address_zip",
"LATITUDE": "lat",
"LONGITUDE": "long",
}
df.rename(columns=orlando_rename_dict, inplace=True)
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_sf_add(df):
sf_rename_dict = {
"Parcel Number": 'parcelID',
'Unit Number': 'address_u',
'Address Number': 'address_n1',
'Street Name': 'address_sn',
'Street Type': 'address_ss',
'ZIP Code': 'address_zip',
'Address': 'address_fa',
#'PIN': 'parcelID',
'Longitude': 'long',
'Latitude': 'lat'
}
df.rename(columns=sf_rename_dict, inplace=True)
df['address_city'] = "San Francisco"
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_sd_add(df):
sd_rename_dict = {
'addrunit': 'address_u',
'addrnmbr': 'address_n1',
'addrpdir': 'address_sd',
'addrname': 'address_sn',
'addrsfx': 'address_ss',
'addrzip': 'address_zip',
'community': 'address_city',
'PIN': 'parcelID',
}
df.rename(columns=sd_rename_dict, inplace=True)
df = df.to_crs(default_crs)
df.crs = default_crs
df['long'] = df.geometry.centroid.x
df['lat'] = df.geometry.centroid.y
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_la_add(df):
la_rename_dict = {
'AIN': 'parcelID',
'UnitName': 'address_u',
'Number': 'address_n1',
'PostType': 'address_ss',
'PreDirAbbr': 'address_sd',
'ZipCode': 'address_zip',
'LegalComm': 'address_city',
}
df.rename(columns=la_rename_dict, inplace=True)
combine_names(df,
name_cols=['PreType', 'StArticle', 'StreetName'],
newCol="address_sn")
df = df.to_crs(default_crs)
df.crs = default_crs
df['long'] = df.geometry.centroid.x
df['lat'] = df.geometry.centroid.y
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df
def clean_int_addresses(df):
df = add_subset_address_cols(df)
df = parallelize_dataframe(df=df, func=clean_parse_parallel, n_cores=2)
return df