Pandas replace values in column

In this article, we will discuss how to replace values in column of pandas dataframe.
[toc]

Using the loc() function to replace values in column of pandas DataFrame

The loc() function is used to access values based on column names and row values. We can use this function to access the required value and provide the new value using the = operator.

For example,

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df.loc[0,'a'] = 'Jack'
print(df)    

Output:

      a   b   c
0  Jack  75  18
1  Mark  92  17
2  Neil  74  18

In the above example, we changed Jay to Jack in column a.

We can use this function to replace values based on some conditions also.

See the following example.

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df.loc[df.b<80,'b'] = 60
print(df)    

Output:

      a   b   c
0   Jay  60  18
1  Mark  92  17
2  Neil  60  18

In the above example, we found all the values below 75 in column b and replaced them with 60.

Further reading:

Get unique values in column of Pandas dataframe

Read more →

Drop rows in Pandas

Read more →

Using the iloc() function to to replace values in column of pandas DataFrame

The iloc() function is similar to the loc() function and can be used to access columns and rows of a DataFrame. It uses the index of the values instead of their labels.

For example,

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df.iloc[0,0] = 'Jack'
print(df)        

Output:

      a   b   c
0  Jack  75  18
1  Mark  92  17
2  Neil  74  18

Using the map() function to replace values of a column in a pandas DataFrame

The map() function can apply some function or collector on all the elements of a Series object or a DataFrame. We can use it to detect and replace values of a column in a DataFrame. We have to specify the old and new values within the function.

For example,

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df['a'] = df['a'].map({'Jay':'Jack', 'Mark':'Mac', 'Neil':'Neil'})
print(df)            

Output:

      a   b   c
0  Jack  75  18
1   Mac  92  17
2  Neil  74  18

Note that we also need to name the element we do not wish to replace otherwise, they will get replaced with the NaN constant.

Using the replace() function to replace values in column of pandas DataFrame

This probably the most straightforward method to replace the values of a column. We can use the replace() function to replace one or more values in a DataFrame.

In the following example, we will replace multiple values with one value.

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df['a'] = df['a'].replace(['Jay','Mark'],'Unknown')
print(df)        

Output:

         a   b   c
0  Unknown  75  18
1  Unknown  92  17
2     Neil  74  18

Similarly, we can replace multiple values with different values for each.

For example,

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df['a'] = df['a'].replace(['Jay','Mark'],['Jack','Mac'])
print(df)        

Output:

      a   b   c
0  Jack  75  18
1   Mac  92  17
2  Neil  74  18    

This function can also be applied to the whole DataFrame.

Using the where() function to replace values in column of pandas DataFrame

The where() function checks the DataFrame to detect some values based on a given condition. We can replace the values which satisfy the given condition with some new value.

See the following example.

import pandas as pd
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df['b'].where(~(df.b < 80), other=60, inplace=True)
print(df)        

Output:

      a   b   c
0   Jay  60  18
1  Mark  92  17
2  Neil  60  18    

The above example replaces all values less than 80 with 60.

Using the numpy.where() function to to replace values in column of pandas DataFrame

The where() function from the numpy module is generally used with arrays only. However, since we need to change the values of a column, we can use this function with a pandas DataFrame also.

This method works similarly to the method discussed previously.

For example,

import pandas as pd
import numpy as np
df = pd.DataFrame([['Jay',75,18],['Mark',92,17],['Neil',74,18]], columns = ['a','b','c'])
df['b'] = np.where((df.b < 80), 60, df.b)
print(df)        

Output:

      a   b   c
0   Jay  60  18
1  Mark  92  17
2  Neil  60  18    

That's all about how to replace values in column of pandas DataFrame.