In [3]:

    

%load_ext autoreload


    

In [4]:

    

%autoreload 2
from geocode_mak import lat_lon_from_mak_names
import pandas as pd


    

In [5]:

    

mak_df = pd.read_csv("/Users/hep/Downloads/19_10_2017_organisations_to_geocode.csv",names=["institutes","n"])


    

In [9]:

    

lat_lon_match_score = lat_lon_from_mak_names(mak_df["institutes"],"~/Downloads/grid20170810/",perfect_only=True)


    

    




100%|██████████| 3979/3979 [00:36<00:00, 110.47it/s]

In [14]:

    

mak_df["longitude"] = [lon for lat,lon,match,score in lat_lon_match_score]
mak_df["latitude"] = [lat for lat,lon,match,score in lat_lon_match_score]
mak_df["match"] = [match for lat,lon,match,score in lat_lon_match_score]
mak_df["score"] = [score for lat,lon,match,score in lat_lon_match_score]


    

In [15]:

    

mak_df.to_csv("/Users/hep/Downloads/19_10_2017_organisations_geocoded_perfectOnly.csv")


    

In [17]:

    

mak_df.tail()


    

    
Out[17]:







  

    

      

      
institutes
      
n
      
longitude
      
latitude
      
match
      
score
    
  
  

    

      
3974
      
department of environment and conservation
      
1
      
-52.721597
      
47.558815
      
Department of Environment and Conservation
      
1.0
    
    

      
3975
      
babol noshirvani university of technology
      
1
      
52.681725
      
36.561561
      
Babol Noshirvani University of Technology
      
1.0
    
    

      
3976
      
united nations industrial development organiza...
      
1
      
16.416660
      
48.235761
      
United Nations Industrial Development Organiza...
      
1.0
    
    

      
3977
      
cgi group
      
1
      
NaN
      
NaN
      
None
      
0.0
    
    

      
3978
      
armasuisse
      
1
      
NaN
      
NaN
      
None
      
0.0
    
  

In [19]:

    

null_long = pd.isnull(mak_df.longitude)
null_lat = pd.isnull(mak_df.latitude)
null_match = pd.isnull(mak_df.match)

print((~(null_long | null_lat)).sum())
print(((null_long | null_lat) & null_match).sum())
print(((null_long | null_lat) & ~null_match).sum())


    

    




2389
1410
180

In [20]:

    

mak_df.to_dict(orient="records")[0]


    

    
Out[20]:





{'institutes': 'united arab emirates university',
 'latitude': 24.198948,
 'longitude': 55.678852,
 'match': 'United Arab Emirates University',
 'n': 37519,
 'score': 1.0}

In [ ]: