using_python_for_data_cleaning.py

# -*- coding: utf-8 -*-
"""Using Python For Data Cleaning.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1ZWWYTVAOj2NveC_JEtCkQDow0CSxrInM
"""

# Load Packages
import pandas as pd

# Load Dataset
df = pd.read_csv("dataset.csv")

df.head()

# Column Consistency
df.columns

df.columns.str.lower()

df.columns = df.columns.str.lower()

# Renaming
df.rename(columns={'full name':'full_name','date of birth':'date_of_birth'})

df.columns

# Rename and Replace Column Names
df.rename(columns={'full name':'full_name','date of birth':'date_of_birth'},inplace=True)

df.columns

# Method 1
df.full_name

df.full_name.str.split(" ")

# Get the First Part of the Split
df.full_name.str.split(" ").str.get(0)

df['firstname'] = df.full_name.str.split(" ").str.get(0)

df.firstname

# Last Name
df['lastname'] = df.full_name.str.split(" ").str.get(1)

df.lastname

# Method 2
df.full_name

df1 = df

# Method 2 Using Expand
df1.full_name.str.split(" ",expand=True)

# Using expand and n = 1 to group all other into one column
df1.full_name.str.split(" ",n=1,expand=True)

df.head(13)

df['income.1']

# Data Type of Column
df['income.1'].dtype

# Replace with Empty
df['income.1'].str.replace("$"," ")

# Replace with Euro
df['income.1'].str.replace("$","Euro")

df.head()

df.salary

# As Boolean
df.salary.str.contains('19')

# Get the values of entire rows
df[df.salary.str.contains('19')]

# Check For Multiple Expression
df.salary.str.contains('19|17')

df.salary.str.contains('19|17',regex=True)

# Using Match to Find An Expression
df.salary.str.match('19')

df.quote

# Find the match for the word Operative
df[df.quote.str.match('Operative')]

# Finding An Index
df.salary.filter(regex='18',axis=0)

# Joining  Column
df.head()

# Method1
df.firstname + df.email

# Method 1
df.firstname +"_"+ df.email

# Method 2
dfall = df[['firstname','email']].apply("_".join,axis=1)

dfall

# Counting Strings in A Column
df.quote

# Using str.count() to count number of space
df.quote.str.count(' ') + 1

# Using str.len() to get length of words in each row of a column
df.quote.str.split().str.len()

# Using str.split().map(len) to get length of words in each row of a column
df.quote.str.split().map(len)

# Using str.split().apply(len) to get length of words in each row of a column
df.quote.str.split().apply(len)

# Get the Total Number of Counts
df.quote.str.split().apply(len).value_counts()

df.head()