import pandas as pd

person = {
    "first": ["Kushagra", "Jane", "John"],
    "last": ["Gupta", "Doe", "Doe"],
    "email": ["Kushagra225@gmail.com", "JaneDoe123@gmail.com", "John@gmail.com"]
}

df = pd.DataFrame(person)

df

df.columns

Index(['first', 'last', 'email'], dtype='object')

#changing column names,, but you need to pass all the column names
df.columns = ['first_name', 'last_name', 'email']

df

#using list comprehension
df.columns = [x.upper() for x in df.columns]

df

df.columns = df.columns.str.replace('_', ' ')

df

#renaming specific column
df.rename(columns={'FIRST NAME': 'first', 'LAST NAME': 'last'})

df

#inplace is used to make save change in original df
df.rename(columns={'FIRST NAME': 'first', 'LAST NAME': 'last'}, inplace=True)

#changing data of row
df.loc[2] = ['John', 'Smith', 'John@gmail.com']

df

df.rename(columns={'EMAIL': 'email'},inplace=True)

#changing specific columns of row
df.loc[2,['first', 'email']] = ['Jack', 'Jack@gmail.com']

df

#you can also use 'at' method to change single value
df.at[2, 'last'] = 'Doe'

df

filt = (df['email'] == 'Jack@gmail.com')
df[filt]

#this will not change last_name as in background it return multiple operations, bcoz value is set on a temporary object that is made
df[filt]['last'] = 'Smith'

/Users/kushagra/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

#you can use loc method with filter for setting up the value
filt = (df['email'] == 'Jack@gmail.com')
df.loc[filt, 'last'] = 'Smith'
df

#changing all emails to lower case
df['email'] = df['email'].str.lower()

df

# apply - used to work with series objects.. when function is applied with series object it gets applied to every value (each value of column), when function is applied with df it gets applied to each series objects (column/row)
# applymap - used to work with df, applies function to each individual value in df
# map - only works with series objects, returns NaN if value is not provided
# replace - only works with series objects, returns previous value for value not provided

# apply
# checking length of email addresses
df['email'].apply(len)

0    21
1    20
2    14
Name: email, dtype: int64

def update_email(email):
    return email.upper()

#we can also use custom function with apply
df['email'].apply(update_email)

0    KUSHAGRA225@GMAIL.COM
1     JANEDOE123@GMAIL.COM
2           JACK@GMAIL.COM
Name: email, dtype: object

#same with lambda function
df['email'].apply(lambda x: x.upper())

0    KUSHAGRA225@GMAIL.COM
1     JANEDOE123@GMAIL.COM
2           JACK@GMAIL.COM
Name: email, dtype: object

print(df['email'].apply(len)) # applying in series objects
print(df.apply(len)) # applying in dataframes same as len(df['email'])

0    21
1    20
2    14
Name: email, dtype: int64
first    3
last     3
email    3
dtype: int64

# applymap
df.applymap(len)

df =df.applymap(str.lower)

#map
#returns NaN if value is not provided
df['first'].map({'kushagra':'kush', 'jane': 'Jane'})

0    kush
1    Jane
2     NaN
Name: first, dtype: object

#replace
#returns previous value for value not provided
df['first'].replace({'kushagra':'kush', 'jane': 'Jane'})

0    kush
1    Jane
2    jack
Name: first, dtype: object

	first	last	email
0	Kushagra	Gupta	Kushagra225@gmail.com
1	Jane	Doe	JaneDoe123@gmail.com
2	John	Doe	John@gmail.com