How am I able to replace duplicates in a dataframe column in python?

Question

say my column is something like this:

trade_signal
buy
buy
buy
buy
sell
sell
sell
sell
buy
buy
buy
sell
sell
buy
sell
buy

I would like to drop the duplicate elements in the column and replace them with NAN or 0 so it would end up with something like:

trade_signal
buy
nan
nan
nan
sell
nan
nan
nan
buy
nan
nan
sell
nan
buy
sell
buy

I am completely unsure of the logic I can use to do this, I think I would forward fill up until the next change in signal with NAN values somehow?

anky · Accepted Answer · 2022-03-21 15:51:43Z

2

Try mask with shift:

df['trade_signal'] = df['trade_signal'].mask(df['trade_signal'].eq(
                                            df['trade_signal'].shift())
                                             )

  trade_signal
0           buy
1           NaN
2           NaN
3           NaN
4          sell
5           NaN
6           NaN
7           NaN
8           buy
9           NaN
10          NaN
11         sell
12          NaN
13          buy
14         sell
15          buy

answered Mar 21, 2022 at 15:51

anky

75.3k11 gold badges46 silver badges76 bronze badges

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

5 Comments

Jason_Leto Over a year ago

thank you this is great, is it possible to fill nans up until the next buy/sell signal? so if it went buy,nan,nan,buy,nan,sell it would end up like buy,nan,nan,nan,nan,sell

anky Over a year ago

@Jason_Leto can you give me an example series dict ? (also try the code once since I think this will handle that)

Jason_Leto Over a year ago

The code doesnt do that, for example df['Trade_Signal'] = pd.Series(np.array(['buy', 'NaN', 'Buy', 'Nan', 'Sell'])) i would like to be all nans between the first buy and last element(sell)

Jason_Leto Over a year ago

Sorry I used that as an example, changed it above. It gives me ['buy'',nan','buy',nan,'sell'] i would like ['buy'',nan''nan,nan,'sell']

anky Over a year ago

@Jason_Leto just do df['trade_signal']=df['trade_signal'].ffill() before trying this code and let me know

Collectives™ on Stack Overflow

How am I able to replace duplicates in a dataframe column in python?

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