Map vs apply vs applymap pandas: Key Differences and Usage

map is used to map values of a Series according to an input correspondence, apply applies a function along an axis of a DataFrame or on a Series, and applymap applies a function elementwise to every cell in a DataFrame. Each serves different scopes: map for Series value mapping, apply for row/column-wise operations, and applymap for element-wise DataFrame transformations.

Quick Comparison

Here is a quick table summarizing the main differences between map, apply, and applymap in pandas.

Function	Data Type	Scope	Input	Output	Typical Use Case
map	Series	Element-wise on Series values	dict, Series, or function	Series	Replace or map Series values
apply	Series or DataFrame	Row-wise or column-wise on DataFrame or element-wise on Series	function	Series or DataFrame	Apply function along axis or on Series
applymap	DataFrame	Element-wise on all DataFrame cells	function	DataFrame	Apply function to each cell in DataFrame

Key Differences

map is designed specifically for Series. It replaces or maps each value in the Series using a dictionary, another Series, or a function. It is not available for DataFrames.

apply is more flexible. For a DataFrame, it applies a function along either rows or columns (axis=0 or axis=1). For a Series, it applies a function element-wise. The output shape can vary depending on the function.

applymap works only on DataFrame objects and applies a function to every single element individually. It is useful for element-wise transformations when you want to change each cell's value.

Code Comparison

Example: Convert a Series of letters to uppercase using map.

python

import pandas as pd

s = pd.Series(['a', 'b', 'c', 'd'])
result = s.map(str.upper)
print(result)

Output

0    A
1    B
2    C
3    D
dtype: object

apply and applymap Equivalent

Equivalent operation using apply on a Series and applymap on a DataFrame.

python

import pandas as pd

# Using apply on Series
s = pd.Series(['a', 'b', 'c', 'd'])
result_apply = s.apply(str.upper)

# Using applymap on DataFrame

df = pd.DataFrame({'col1': ['a', 'b'], 'col2': ['c', 'd']})
result_applymap = df.applymap(str.upper)

print(result_apply)
print(result_applymap)

Output

0    A
1    B
2    C
3    D
dtype: object
  col1 col2
0    A    C
1    B    D

When to Use Which

Choose map when you want to replace or map values in a Series using a dictionary or function. Use apply when you need to apply a function along rows or columns of a DataFrame or element-wise on a Series. Use applymap when you want to apply a function to every single element in a DataFrame individually.

In short, map is for simple value mapping on Series, apply is for flexible row/column or Series-wise operations, and applymap is for element-wise DataFrame transformations.

Key Takeaways

map is for mapping or replacing values in a Series using dict, Series, or function.

apply applies functions along DataFrame rows/columns or element-wise on Series.

applymap applies a function element-wise to every cell in a DataFrame.

Use map for simple value substitution, apply for flexible axis-wise operations, and applymap for element-wise DataFrame changes.