Removing duplicate rows from a csv file using a python script

Posted on

Question :

Removing duplicate rows from a csv file using a python script

Goal

I have downloaded a CSV file from hotmail, but it has a lot of duplicates in it. These duplicates are complete copies and I don’t know why my phone created them.

I want to get rid of the duplicates.

Approach

Write a python script to remove duplicates.

Technical specification

Windows XP SP 3
Python 2.7
CSV file with 400 contacts

Asked By: IcyFlame

||

Answer #1:

UPDATE: 2016

If you are happy to use the helpful more_itertools external library:

from more_itertools import unique_everseen
with open('1.csv','r') as f, open('2.csv','w') as out_file:
    out_file.writelines(unique_everseen(f))

A more efficient version of @IcyFlame’s solution

with open('1.csv','r') as in_file, open('2.csv','w') as out_file:
    seen = set() # set for fast O(1) amortized lookup
    for line in in_file:
        if line in seen: continue # skip duplicate

        seen.add(line)
        out_file.write(line)

To edit the same file in-place you could use this

import fileinput
seen = set() # set for fast O(1) amortized lookup
for line in fileinput.FileInput('1.csv', inplace=1):
    if line in seen: continue # skip duplicate

    seen.add(line)
    print line, # standard output is now redirected to the file
Answered By: jamylak

Answer #2:

you can achieve deduplicaiton efficiently using Pandas:

import pandas as pd
file_name = "my_file_with_dupes.csv"
file_name_output = "my_file_without_dupes.csv"

df = pd.read_csv(file_name, sep="t or ,")

# Notes:
# - the `subset=None` means that every column is used 
#    to determine if two rows are different; to change that specify
#    the columns as an array
# - the `inplace=True` means that the data structure is changed and
#   the duplicate rows are gone  
df.drop_duplicates(subset=None, inplace=True)

# Write the results to a different file
df.to_csv(file_name_output, index=False)
Answered By: Andrei Sura

Answer #3:

You can use the following script:

pre-condition:

  1. 1.csv is the file that consists the duplicates
  2. 2.csv is the output file that will be devoid of the duplicates once this script is executed.

code



inFile = open('1.csv','r')

outFile = open('2.csv','w')

listLines = []

for line in inFile:

    if line in listLines:
        continue

    else:
        outFile.write(line)
        listLines.append(line)

outFile.close()

inFile.close()


Algorithm Explanation

Here, what I am doing is:

  1. opening a file in the read mode. This is the file that has the duplicates.
  2. Then in a loop that runs till the file is over, we check if the line
    has already encountered.
  3. If it has been encountered than we don’t write it to the output file.
  4. If not we will write it to the output file and add it to the list of records that have been encountered already
Answered By: IcyFlame

Answer #4:

I know this is long settled, but I have had a closely related problem whereby I was to remove duplicates based on one column. The input csv file was quite large to be opened on my pc by MS Excel/Libre Office Calc/Google Sheets; 147MB with about 2.5 million records. Since I did not want to install a whole external library for such a simple thing, I wrote the python script below to do the job in less than 5 minutes. I didn’t focus on optimization, but I believe it can be optimized to run faster and more efficient for even bigger files. The algorithm is similar to @IcyFlame above, except that I am removing duplicates based on a column (‘CCC’) instead of whole row/line.

import csv

with open('results.csv', 'r') as infile, open('unique_ccc.csv', 'a') as outfile:
    # this list will hold unique ccc numbers,
    ccc_numbers = []
    # read input file into a dictionary, there were some null bytes in the infile
    results = csv.DictReader(infile)
    writer = csv.writer(outfile)

    # write column headers to output file
    writer.writerow(
        ['ID', 'CCC', 'MFLCode', 'DateCollected', 'DateTested', 'Result', 'Justification']
    )
    for result in results:
        ccc_number = result.get('CCC')
        # if value already exists in the list, skip writing it whole row to output file
        if ccc_number in ccc_numbers:
            continue
        writer.writerow([
            result.get('ID'),
            ccc_number,
            result.get('MFLCode'),
            result.get('datecollected'),
            result.get('DateTested'),
            result.get('Result'),
            result.get('Justification')
        ])

        # add the value to the list to so as to be skipped subsequently
        ccc_numbers.append(ccc_number)
Answered By: Ongati Felix

Answer #5:

A more efficient version of @jamylak’s solution: (with one less instruction)

with open('1.csv','r') as in_file, open('2.csv','w') as out_file:
    seen = set() # set for fast O(1) amortized lookup
    for line in in_file:
        if line not in seen: 
            seen.add(line)
            out_file.write(line)

To edit the same file in-place you could use this

import fileinput
seen = set() # set for fast O(1) amortized lookup
for line in fileinput.FileInput('1.csv', inplace=1):
    if line not in seen:
        seen.add(line)
        print line, # standard output is now redirected to the file   
Answered By: Ahmed Abdelkafi

Answer #6:

You can do using pandas library in jupyter notebook or relevant IDE, I m importing pandas to jupyter notebook and reading the csv file

Then sort the values,accordingly by which parameters duplicates are present, since I have defined two attributes first it will sort by time, then by latitude

Then remove duplicates as present in time column or column relevant as per you

Then i store the duplicates removed and sorted file as gps_sorted

import pandas as pd
stock=pd.read_csv("C:/Users/Donuts/GPS Trajectory/go_track_trackspoints.csv")
stock2=stock.sort_values(["time","latitude"],ascending=True)
stock2.drop_duplicates(subset=['time'])
stock2.to_csv("C:/Users/Donuts/gps_sorted.csv",)

Hope this helps

Answered By: Dulangi_Kanchana

Leave a Reply

Your email address will not be published.