Why The Empty Rows aren't getting Deleted (Pandas Dataframe)?

Question

EDIT : Sorry, I got confused rows with Columns,(noob here)

I have a CSV , where there are 3 columns, heading, image and description .

Some of the items in description is Empty , and I want to delete the complete column if description is empty or has length < 1

Firstly, I got all columns which has length of description < 1 using this Code :

for i in df['description']:
if len(i) < 1 :
    print('Empty')

Output :

Empty
Empty

This means there are 2 columns which has description of length < 1 , Now i try to remove them :

for i in df['description']:
if len(i) < 1 :
    df.drop(i, inplace=True, axis=1)

But, still there are those 2 columns with Empty data, how to remove them, What is wrong in my code ? Please Guide

also tried :

df = df.drop(df.columns[[i]], axis=1)

but, nothing works

crissal · Accepted Answer · 2021-01-10 14:35:33Z

2

I think there is a little of confusion. You say you want to remove the entire column, but this means deleting the description column; instead, I think you're talking about the row with empty description.

If this is your case, then you just need to cycle over your dataframe (with an index too) and check if the description is null or empty; if that's true, you can drop the row in-place (index=N means i, axis=0).

for i, description in zip(df.index, df['description']):
    if not description:  # True if None and empty string
        df.drop(index=i, inplace=True)

This is an example of what you're getting.

before drop:
  heading image description
0     foo     a        None
1     bar     b            
2     baz     c   some text

after drop:
  heading image description
2     baz     c   some text

If, however, you need to remove the entire column if there is a None/empty string, then you need to check first if there are values that match your search, and if yes, remove that column.

df.applymap() applies a function to all DataFrame, then you keep only the description column, and you got a bool Series. With any() you check if at least one element is True (None or empty string). If yes, drop all the description column.

if any(df.applymap(lambda x: not x)['description']):
    df.drop(columns='description', inplace=True)

Again, here's an example of result.

before drop:
  heading image description
0     foo     a        None
1     bar     b            
2     baz     c   some text

after drop:
  heading image
0     foo     a
1     bar     b
2     baz     c

answered Jan 10, 2021 at 14:35

crissal

2,64710 silver badges27 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

Sainita Over a year ago

Sorry, I got confused rows with Columns,(noob here) . I wanted to delete the row and not column

Sainita Over a year ago

Following Your Code , I tried this : for i in df['description']: if len(i) < 1 : df.drop(index=i, inplace=True) , still it has the items. not deleted

crissal Over a year ago

I gave you the right answer with an example of result. By the way, you're not cycling over index (i), but over values (None, '', 'some text', etc)

Sainita Over a year ago

How to write correct One Sir, please Guide

crissal Over a year ago

Just check the first code block in my answer, that's the correct for loop

Collectives™ on Stack Overflow

Why The Empty Rows aren't getting Deleted (Pandas Dataframe)?

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related