labels not contained in axis error - pandas dataframe

Question

I have a Dataframe named df which looks like -

pageno     entity          code         rawentity 
17727425   SAUDI           CBCNTRY      saudi 
17727425   GARRA           DRWRNAME     garra
17727425   PO BOX          RBCNTRY      po box 
17727425   NEW ZEALAND     DRWRCNTRY    new zealand

I also have a country list containing names of countries.It is of type 'list'.

I'm trying to keep only those value whose code is either - CBCNTRY or RBCNTRY or DRWRCNTRY and the entity should be in the countrylist.

The code that I have written is -

        for row in df.itertuples():

            if(row.code in ['DRWRCNTRY', 'RBCNTRY', 'CBCNTRY']):
                if(row.entity not in countrylist):
                    df.drop((row.index), inplace=True)

But I'm getting the following error -

Error is: labels [<built-in method index of Pandas object at 0x0000020A1BCE4EB8>] not contained in axis

I just want to know why my approach is wrong and is there any thing better I can do apart from this method.

I have searched this error but I'm unable to get a satisfactory answer.

You are removing rows as you iterate over your dataframe. You should never do this. See @chuni0r's answer. — jpp
– jpp, Commented Jul 9, 2018 at 16:02

chuni0r · Accepted Answer · 2018-07-10 14:34:53Z

1

As suggested by Gerardo, use pd.isin and combine your expressions using boolean operators:

countrylist = ['SAUDI']
codelist = ['DRWRCNTRY', 'RBCNTRY', 'CBCNTRY']
df = df[(df['code'].isin(codelist)) & (df['entity'].isin(countrylist))]

results in

     pageno entity     code rawentity
0  17727425  SAUDI  CBCNTRY     saudi

edited Jul 10, 2018 at 14:34

answered Jul 9, 2018 at 15:33

chuni0r

1734 bronze badges

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

Comments

Yuca · Accepted Answer · 2018-07-09 15:11:30Z

1

When you're working with series you can use pd.isin. For example, what you want can be achieved by doing:

df = df[df['code'].isin(['DRWRCNTRY', 'RBCNTRY', 'CBCNTRY'])]

answered Jul 9, 2018 at 15:11

Yuca

6,1114 gold badges26 silver badges45 bronze badges

Comments

user96564 · Accepted Answer · 2018-07-09 15:23:51Z

1

You can use np.where and isin

suppose you have list called countrylist that contains SAUDI

countrylist = ['SAUDI']
df['code'] = np.where((df['code'] == 'CBCNTRY')| (df['code'] == 'RBCNTRY') | (df['code'] == 'DRWRCNTRY'),
                      df['code'], np.nan
                      )

df['code'] = np.where(df['entity'].isin(countrylist), df['code'], np.nan)

df.dropna(how='any', inplace= True)

print(df)

  pageno entity     code rawentity
17727425  SAUDI  CBCNTRY     saudi

answered Jul 9, 2018 at 15:23

user96564

1,6175 gold badges27 silver badges47 bronze badges

Collectives™ on Stack Overflow

labels not contained in axis error - pandas dataframe

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related