apply, applymap, map

Reference:

Difference between apply, applymap, and map

the following table is copied from this stackoverflow page

png

When to use apply

Note: the following are copied from this stackoverflow page

png

The following shows the performances of different approaches to converting a dataframe of 2 date columns stored in object (i.e. string) format into datetime format.

png

import pandas as pd
print(f'pandas version: {pd.__version__}')

pandas version: 1.3.4

##construct a dataframe
df = pd.DataFrame(data = {'abbr':['A', 'B', 'C'], 'val':[1, 2, 3]})
df
abbr val
0 A 1
1 B 2
2 C 3 
#use map
map_dict = {'A':'Apple', 'B':'Banana', 'C': 'Cherry'}
df['fullname']=df['abbr'].map(map_dict)
df
abbr val fullname
0 A 1 Apple
1 B 2 Banana
2 C 3 Cherry 
#use apply
df['name_len']=df['fullname'].apply(lambda x:len(x))
df
abbr val fullname name_len
0 A 1 Apple 5
1 B 2 Banana 6
2 C 3 Cherry 6 
#use apply
df['print_out'] = df[['abbr', 'val']].apply(lambda x:f'{x["abbr"]}:{x["val"]}', axis=1)
df
abbr val fullname name_len print_out
0 A 1 Apple 5 A:1
1 B 2 Banana 6 B:2
2 C 3 Cherry 6 C:3 
#use applymap
df['val_pow2']=df[['val']].applymap(lambda x: x**2)['val']
df
abbr val fullname name_len print_out val_pow2
0 A 1 Apple 5 A:1 1
1 B 2 Banana 6 B:2 4
2 C 3 Cherry 6 C:3 9