CODEWITHSUNDEEP
pythonpandasdataframesplitrename
Stacey
Stacey

Reputation: 5107

Change dataframe row names

I have a df which looks like:

BBG.LON.123.S_CAR_ADJ_DPS    343.94325
BBG.LON.436.S_CAR_ADJ_DPS    236.51530

I am trying to rename the row names (removing the '_CAR_ADJ_DPS' element of each row name and rename the column 'id' so my resulting df looks like:

           id
BBG.LON.123.S   343.94325
BBG.LON.436.S   236.51530

I have tried using the following line without success:

pd.DataFrame(pd.Series(np.unique([row.split('_')[0] for row in df.rows]), name='id'))

What can I try next?

Upvotes: 4

Views: 4581

Answers (2)

Andy Hayden
Andy Hayden

Reputation: 375545

You may also be interested in str.extract to pull out the entries as columns:

In [11]: df[0].str.extract('(?P<A>.*)\.(?P<B>.*)\.(?P<C>\d+)\.(?P<D>.)_.*', expand=True)
Out[11]:
     A    B    C  D
0  BBG  LON  123  S
1  BBG  LON  436  S

Upvotes: 2

jezrael
jezrael

Reputation: 862761

I think you can use str.split with rename_axis (new in pandas 0.18.0):

print (df)
                                   a
BBG.LON.123.S_CAR_ADJ_DPS  343.94325
BBG.LON.436.S_CAR_ADJ_DPS  236.51530

df.index = df.index.str.split('_').str[0]
df = df.rename_axis('id')
#if use pandas bellow 0.18.0
#df.index.name = 'id'

print (df)
                       a
id                      
BBG.LON.123.S  343.94325
BBG.LON.436.S  236.51530

Upvotes: 3

Related Questions