Ignore nan cells when performing a pandas lambda map

March 10, 2022

I have the following dataframe and mapping dictionary:

df = pd.DataFrame({ 
     'Name': ['Jim','Jack','Jim','Jack','Jim','Jack','Mick','Mick'],
    'Day':['Mon','Tue','Mon',np.nan,'Sat','Sun','Tue',np.nan],
    'Value':[10,20,30,40,50,60,70,80],
    })


values = {
    'Mon':['Monday','M']
    ,'Tue':['Tuesday','T']
    ,'Wed':['Wednesday','W']
    ,'Thu':['Thursday','TH']
    ,'Fri':['Friday','F']
    ,'Sat':['Saturday','SA']
    ,'Sun':['Sunday','SU']
}
df

as you can see there are some nan values in the Day column.

I am trying to add the following column:

df['Day_Name'] = 'The name of the person is ' + df['Name'] + ' and the day of the week is ' + df['Day'].map(lambda x: values[x][0])

but i am getting a nan error due to the 2 nan cells present. Is there anything i can skip over the blank cells (and return blank) for those particular rows in the new df['Day_Name'] column? Thanks!

>Solution :

Use dict.get with specify list with empty space for possible after selection first value get empty value:

s = df['Day'].map(lambda x: values.get(x, [''])[0])
df['Day_Name'] = 'The name of the person is ' + df['Name'] + ' and the day of the week is ' + s
print (df)
   Name  Day  Value                                           Day_Name
0   Jim  Mon     10  The name of the person is Jim and the day of t...
1  Jack  Tue     20  The name of the person is Jack and the day of ...
2   Jim  Mon     30  The name of the person is Jim and the day of t...
3  Jack  NaN     40  The name of the person is Jack and the day of ...
4   Jim  Sat     50  The name of the person is Jim and the day of t...
5  Jack  Sun     60  The name of the person is Jack and the day of ...
6  Mick  Tue     70  The name of the person is Mick and the day of ...
7  Mick  NaN     80  The name of the person is Mick and the day of ...