如何使用函数从dataframe列获取第一个和最后一个值

Question

I need to get the first value and last values for the following using a function similar to the way I'm getting the max() and min() values from a Pandas DF column. So this part of my code works perfectly to give me the max and min:

day_high = day_session['High'].rolling(window=1,freq='D').max()
day_low = day_session['Low'].rolling(window=1,freq='D').min()

How do I get the first and last values respectively here using a function?

day_open = day_session.Open.rolling(window=1,freq='D'). ? (what is the 'first' function?)

day_close = day_session.Last.rolling(window=1,freq='D'). ? (what is the 'last' function?)

I already tried .first() and this gives error output ' Rolling' object has no attribute 'first' and same for .last()

This is an example of the data Im using:

data['Last']
Out[96]:
Timestamp
2013-03-07 09:30:00    1440.00
2013-03-07 09:31:00    1439.75
2013-03-07 09:32:00    1440.50
2013-03-07 09:33:00    1441.00
2013-03-07 09:34:00    1441.50
2013-03-07 09:35:00    1441.75
2013-03-07 09:36:00    1440.75
2013-03-07 09:37:00    1440.50
2013-03-07 09:38:00    1440.75
2013-03-07 09:39:00    1440.00

This code is used for the rolling window and works fine for the max() Min() above:

daystart = '9:30'
dayend = '16:14:59'

day_session = data.between_time(daystart,dayend, include_start=True, include_end=True)

I am using concat to utilise this data with other series data in a df called all_vals

all_vals = pd.concat([day_open,day_close,day_high,day_low,day_range,IB_high,IB_low,IB_Range,upper_quartile,lower_quartile,end_session_high,end_session_low], axis=1) 

thanks

Answer 1

select the first and last value of a particular column of a data frame

df['column_name'].iloc[0] # first element 
df['column_name'].iloc[-1] # last element 

Answer 2

Have you tried to resample using ohlc ?

df = pd.DataFrame(
    {'price': [100, 101, 102, 103]},
    index=pd.to_datetime(['2016-07-08 10:00', '2016-07-08 10:00', '2016-07-08 10:00', '2016-07-08 10:00']))

>>> df.resample('D', how={'price': 'ohlc'})
           price                
            open high  low close
2016-07-08   100  103  100   103

You should generally post questions with some sample data and expected results.

EDIT Given your sample data, you can do this:

df.resample('D', how={'Last': 'ohlc'})
            Last                        
            open     high      low close
Timestamp                               
2013-03-07  1440  1441.75  1439.75  1440