CODEWITHSUNDEEP
pythonpandasdataframedata-analysis
Sunmola Oluwaseun
Sunmola Oluwaseun

Reputation: 11

Finding the number of occurrences of a specific string in a column

I'm trying to count the number of words that have the string: "hanger" from the column "Description". So I defined a function:

def hanger_count(title): 
if 'hanger' in title.lower().split():
    return True
else:
    return False

Which seemed to be working correctly when I tested it with a string. But when I attempted to run the function through the data column, using the function:

ecomm['Description'].apply(hangercount)

I received an error back:

AttributeError: 'float' object has no attribute 'lower'

I think the issue is that python is seeing some of the rows in the column as objects rather than strings, is there any way I can convert it?

What do you think I'm doing wrong?

Upvotes: 1

Views: 130

Answers (2)

GSJ
GSJ

Reputation: 41

You appear to have mixed data types in your column, and since lower() is only a method for strings, you are getting an error when pandas attempts to call the function on a numeric value (in this case a float).

This quick tweak might work for you:

def hanger_count(title): 
    if 'hanger' in str(title).lower().split():
        return True
    else:
        return False

Upvotes: 4

rahlf23
rahlf23

Reputation: 9019

You can just do the following using str.contains() and count():

df[df['Description'].str.contains('hanger', case=False, na=False)].count()

Upvotes: 3

Related Questions