Tuan Anh Hoang-Vu hvtuananh

## cities.json
[
    {
        "city": "New York",
        "growth_from_2000_to_2013": "4.8%",
        "latitude": 40.7127837,
        "longitude": -74.0059413,
        "population": "8405837",
        "rank": "1",
        "state": "New York"
    },

## keybase.md

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                hvtuananh
                / keybase.md
            
            
              Created
              October 10, 2016 23:18
            
          
    Keybase proof

I hereby claim:

I am hvtuananh on github.
I am hvtuananh (https://keybase.io/hvtuananh) on keybase.
I have a public key whose fingerprint is 4D67 56F5 E786 100E 244F  B967 6073 1E59 AE10 2257

To claim this, I am signing this object:

  
## join.py
from multiprocessing import Pool, cpu_count
import sys
import csv
from utils import Entity
import pickle

file1 = sys.argv[1]
fields1 = map(int, sys.argv[2].split(','))
file2 = sys.argv[3]
fields2 = map(int, sys.argv[4].split(','))

## entity_similarity.py
import re
from difflib import SequenceMatcher


pattern=re.compile("[^\w']")


def gen_signature(string):
    string = string.lower()
    string = pattern.sub(' ', string)

## us_address_abbreviations.txt
ALLEE,ALLEY
ALLEY,ALLEY
ALLY,ALLEY
ALY,ALLEY
ANEX,ANNEX
ANNEX,ANNEX
ANNX,ANNEX
ANX,ANNEX
ARC,ARCADE
ARCADE,ARCADE

## unicode_csv.py
#http://docs.python.org/2.7/library/csv.html

import csv, codecs, cStringIO

class UTF8Recoder:
    """
    Iterator that reads an encoded stream and reencodes the input to UTF-8
    """
    def __init__(self, f, encoding):
        self.reader = codecs.getreader(encoding)(f)
	[
	{
	"city": "New York",
	"growth_from_2000_to_2013": "4.8%",
	"latitude": 40.7127837,
	"longitude": -74.0059413,
	"population": "8405837",
	"rank": "1",
	"state": "New York"
	},
	from multiprocessing import Pool, cpu_count
	import sys
	import csv
	from utils import Entity
	import pickle

	file1 = sys.argv[1]
	fields1 = map(int, sys.argv[2].split(','))
	file2 = sys.argv[3]
	fields2 = map(int, sys.argv[4].split(','))
	import re
	from difflib import SequenceMatcher


	pattern=re.compile("[^\w']")


	def gen_signature(string):
	string = string.lower()
	string = pattern.sub(' ', string)
	ALLEE,ALLEY
	ALLEY,ALLEY
	ALLY,ALLEY
	ALY,ALLEY
	ANEX,ANNEX
	ANNEX,ANNEX
	ANNX,ANNEX
	ANX,ANNEX
	ARC,ARCADE
	ARCADE,ARCADE
	#http://docs.python.org/2.7/library/csv.html

	import csv, codecs, cStringIO

	class UTF8Recoder:
	"""
	Iterator that reads an encoded stream and reencodes the input to UTF-8
	"""
	def __init__(self, f, encoding):
	self.reader = codecs.getreader(encoding)(f)