使用 Python 遍历与列匹配的日期

Question

I have a table file with 366 columns. From these, one represents the targeted date and the other 365 represents the temperature of one day of the year. I added a new column named “temperature” and I would like to fill in this column with the corresponding temperature value of my date field. Hard to explain, please look at the example:

DateFin   b1_T_M01  b2_T_M01 Temperature
27-01-18    5.6      3.8   
06-01-18    5.6      4.2
02-01-18    6.3      4.6        4.6
01-01-18    3.2      5.9        3.2

From b x _TM x , the b refers to the day and M to the month. So for example “b1_T_M01” refers to the temperature value of day one of the month one (first of January). Said that I would like to reduce the number of dimensions and only to keep the information on the day of my interest. As an example, for date 27-01-18 I would need the band 27 from month 01 (b27_TM01). I have no idea of how to iterate using different dimensions, I am frozen. Does anyone have a clue of how to do that in an iteratively way using python? Thank you very much for you attention!

Answer 1

You can build a string out of your date column, to get the right value:

df['DateFin'] = pd.to_datetime(df['DateFin'], dayfirst=True)

def my_date_string(date):
    return 'b' + str(date.day) + '_T_M' + '{:02d}'.format(date.month)


df['Temperature'] = df.apply(lambda row:
         row[my_date_string(row['DateFin'])]
         if my_date_string(row['DateFin']) in df.columns
         else '', axis=1)
print(df)

     DateFin  b1_T_M01  b2_T_M01 Temperature
0 2018-01-27       5.6       3.8            
1 2018-01-06       5.6       4.2            
2 2018-01-02       6.3       4.6         4.6
3 2018-01-01       3.2       5.9         3.2