In [5]:
import pandas as pd
import numpy as np
from map_functions import *
df = pd.read_csv('data/locations (no_lat-lng)_2017.csv')
df.tail()
Out[5]:
In [6]:
# Add columns for origin and destination payloads (for Google API)
payload_pre = 'https://maps.googleapis.com/maps/api/geocode/json?address='
df['Payload_Orig'] = df['Name_Orig'].str.replace(', ','+').str.replace(' ','+')
df['Payload_Des'] = df['Name_Des'].str.replace(', ','+').str.replace(' ','+')
df['Payload_Orig'] = payload_pre + df['Payload_Orig']
df['Payload_Des'] = payload_pre + df['Payload_Des']
df['Distance (nm)'] = 0
df['Distance (mi)'] = 0
df['Distance (km)'] = 0
df.tail()
Out[6]:
I round the decimal degrees and distance measurements for readability. Pick what to round decimal degrees to based on the accuracy required. For our purposes, 3 decimal places should suffice.
This table from Wikipedia gives the precision of decimal degrees.
| Decimal Degrees | Identifiable | Distance At Equator |
|---|---|---|
| 1 | country or large region | 111.32 km |
| 0.1 | large city or district | 11.132 km |
| 0.01 | town or village | 1.1132 km |
| 0.001 | neighborhood, street | 111.32 m |
| 0.0001 | individual street, land parcel | 11.132 m |
| 0.00001 | individual trees | 1.1132 m |
| 0.000001 | individual humans | 111.32 mm |
| 0.0000001 | practical limit of commercial surveying | 11.132 mm |
| 0.00000001 | specialized surveying (e.g. tectonic plate mapping) | 1.1132 mm |
In [7]:
# Find latitude and longitude of each location, and distance between
round_coord = 3
round_dist = 1
print('Completed...')
for row in range(0,df.shape[0]):
try:
# Read payloads to pass to Google API
payload_orig = df.iloc[row,6]
payload_des = df.iloc[row,7]
# Request and write latitude and longitude
lat_orig, lng_orig = get_lat_lng(payload_orig)
lat_des, lng_des = get_lat_lng(payload_des)
df.ix[row,'Lat_Orig'] = round(lat_orig,round_coord)
df.ix[row,'Lng_Orig'] = round(lng_orig,round_coord)
df.ix[row,'Lat_Des'] = round(lat_des,round_coord)
df.ix[row,'Lng_Des'] = round(lng_des,round_coord)
# Calculate and write distance
nm, mi, km = get_distance(lat_orig, lng_orig, lat_des, lng_des)
df.ix[row,'Distance (nm)'] = round(nm,round_dist)
df.ix[row,'Distance (mi)'] = round(mi,round_dist)
df.ix[row,'Distance (km)'] = round(km,round_dist)
print(' - '+ str(round(nm,round_dist))+' nm '+df.iloc[row,0]+' to '+df.iloc[row,3])
except:
print('\tSKIPPING: '+df.iloc[row,0]+' to '+df.iloc[row,3])
print('\t'+payload_orig)
print('\t'+payload_des)
df.tail()
Out[7]:
In [8]:
# Save df to csv
header = ['Name_Orig','Lat_Orig','Lng_Orig','Name_Des','Lat_Des','Lng_Des',
'Distance (nm)','Distance (mi)','Distance (km)']
df.to_csv('data/locations.csv', columns=header, index=False)
In [ ]: