CODEWITHSUNDEEP
pythonpandasdataframelambda
Christine
Christine

Reputation: 33

How to simultaneously apply function with if statement to multiple dataframe columns

I have a dataframe consisting of numeric strings and/or empty strings in multiple columns, and I would like to convert these columns from strings to "int" datatype. Before doing so, I want to convert the empty strings to "-1" (either the int or the string version of -1; it does not matter).

I am trying to simultaneously apply a lambda function to multiple columns to convert the empty string, but am getting an error "'The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().', 'occurred at index Temperature(F)'"

I am posting a dummy example below of what I'm trying to do with my actual dataframe, but it is not working. Of course there is a workaround of iterating through each column in a "for" loop, but I suspect there is a cleaner solution.

df = pd.DataFrame({'Temperature(F)':['30','40',''],'Gust':['','5','10']})
numericCols = ['Temperature(F)','Gust']
df[numericCols]=fTable[numericCols].apply(lambda x:-1 if x=='' else x)
df[numericCols] = fTable[numericCols].astype('int')
'''

As described, I get the error message "'The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().', 'occurred at index Temperature(F)'" when I run this.

Upvotes: 1

Views: 45

Answers (1)

user2314737
user2314737

Reputation: 29307

In one line not using apply

df[numericCols].apply(pd.to_numeric, errors='coerce').fillna(value=-1)
# Out:
#    Temperature(F)  Gust
# 0            30.0  -1.0
# 1            40.0   5.0
# 2            -1.0  10.0

Upvotes: 1

Related Questions