CODEWITHSUNDEEP
pythonpandasdataframejupyter-notebookmultiple-columns
BusterTheFatCat
BusterTheFatCat

Reputation: 57

Dropping range of columns from DataFrame in Pandas

I have a dataframe whose columns I want to drop can be listed by using df._get_numeric_data().

If I write 

for df.column in df._get_numeric_data(): 
    print(df.column)

I will get a list of all the columns in type string I want to drop.

Unnamed: 2
Unnamed: 4
Unnamed: 6
Unnamed: 8
Unnamed: 10
Unnamed: 12
Unnamed: 13
Unnamed: 15
Unnamed: 19
Unnamed: 21
Unnamed: 22
Unnamed: 24
Unnamed: 26
Unnamed: 28
Week Ending
Unnamed: 31
Unnamed: 34
Unnamed: 35
Unnamed: 37
Unnamed: 42
Unnamed: 47

I'm trying to figure out how to drop all those columns in one go. I tried 

for df.column in df._get_numeric_data():
    df.drop(df.column)

But get an error:

KeyError: "['Unnamed: 2'] not found in axis"

I tried a few other things, and think I'm not far off, but something about the df.drop(df.column) isn't correct. Any suggestions?

Upvotes: 1

Views: 405

Answers (1)

BENY
BENY

Reputation: 323386

You should add axis=1, drop default is to drop the index 

for df.column in df._get_numeric_data():
    df.drop(df.column,axis=1)

Also no need for loop

df=df.drop(df._get_numeric_data().column,axis=1)

Upvotes: 4

Related Questions