CODEWITHSUNDEEP
pandas
PeterL
PeterL

Reputation: 515

How to rename Pandas dataframe with unicode character

I have Pandas Dataframe that comes from Excel import:

Trade   Symbol  Type    Signal  Date    Price   Contracts   Profit (¤)  Run-up (¤)  Drawdown (¤)
0   1   FOXA    EntryLong   Buy 2000-10-18  34.69   200 0   0   0
1   NaN NaN ExitLong    Sell    2000-10-18  34.69   200 NaN NaN NaN
2   2   TXN EntryLong   Buy 2000-10-18  36.88   200 1774    1900    0
3   NaN NaN ExitLong    Sell    2000-10-19  45.75   200 NaN NaN NaN

I need to rename "Profit (¤)" to "Profit" but cannot get rid of the "¤" sign.

I am using:

trades.rename(columns={'Trade #':'Trade','Symbol Name':'Symbol','Profit (¤)':'Profit','Run-up (¤)':'Runup','Drawdown (¤)':'Drawdown'}, inplace=True)

which does not work.

list(trades)

gets:

['Trade',
 'Symbol',
 u'Type',
 u'Signal',
 u'Date',
 u'Price',
 u'Contracts',
 u'Profit (\xa4)',
 u'Run-up (\xa4)',
 u'Drawdown (\xa4)']

Any tip how to rename the colums with the special character? Thank you.

Upvotes: 1

Views: 777

Answers (1)

EdChum
EdChum

Reputation: 394101

It looks like all you're doing is just dropping the erroneous additional character, as it's delimited by a space you can just overwrite the column names by splitting on the space:

In [104]:
df = pd.DataFrame(columns=['Trade',
 'Symbol',
 u'Type',
 u'Signal',
 u'Date',
 u'Price',
 u'Contracts',
 u'Profit (\xa4)',
 u'Run-up (\xa4)',
 u'Drawdown (\xa4)'])
df

Out[104]:
Empty DataFrame
Columns: [Trade, Symbol, Type, Signal, Date, Price, Contracts, Profit (¤), Run-up (¤), Drawdown (¤)]
Index: []

In [105]:   
df.columns.str.split().str[0]

Out[105]:
Index(['Trade', 'Symbol', 'Type', 'Signal', 'Date', 'Price', 'Contracts',
       'Profit', 'Run-up', 'Drawdown'],
      dtype='object')

So the following should work for you:

trades.columns = trades.columns.str.split().str[0]

Upvotes: 2

Related Questions