CODEWITHSUNDEEP
pythonpython-3.xpandasdatetime
learning_python
learning_python

Reputation: 111

Calculate total number of values per day with pandas

I have the following data frame

       UNIT  C/A   DATETIME                TOTAL COUNTs
       R248  HOO7  2018-03-03 03:00:00     139.0
                   2018-03-03 07:00:00     927.0
                   2018-03-04 11:00:00    1946.0
                   2018-03-05 07:00:00    1330.0
                   2018-03-05 11:00:00    2358.0

I want to calculate the sum of 'TOTAL COUNTS' for each day.So I want the dataframe to look like this:

        UNIT  C/A   DATETIME     TOTAL COUNTs
       R248  HOO7  2018-03-03    1066
                   2018-03-04    1946
                   2018-03-05    3688

I read the I have to set the index of this dataframe to 'datetime' index and then use .resample, but I am not sure how to do so.

Upvotes: 3

Views: 7142

Answers (2)

jezrael
jezrael

Reputation: 863801

Use if 3 levels MultiIndex use floor or date and aggregate sum:

dates = df.index.get_level_values('DATETIME').floor('D')

Or:

dates = pd.Series(df.index.get_level_values('DATETIME').date, name='DATETIME', index=df.index)

If column DATETIME:

dates = df['DATETIME'.dt.floor('D')
dates = df['DATETIME'].dt.date

df = df.groupby(['UNIT','C/A', dates]).sum().reset_index()
print (df)
   UNIT   C/A   DATETIME  TOTAL COUNTs
0  R248  HOO7 2018-03-03        1066.0
1  R248  HOO7 2018-03-04        1946.0
2  R248  HOO7 2018-03-05        3688.0

Solution with resample:

df = df.set_index('DATETIME').groupby(['UNIT','C/A']).resample('D').sum().reset_index()

Or:

df = df.groupby(['UNIT','C/A', pd.Grouper(key='DATETIME', freq='D')]).sum().reset_index()

print (df)
   UNIT   C/A   DATETIME  TOTAL COUNTs
0  R248  HOO7 2018-03-03        1066.0
1  R248  HOO7 2018-03-04        1946.0
2  R248  HOO7 2018-03-05        3688.0

Upvotes: 3

petezurich
petezurich

Reputation: 10224

Try this:

df.groupby(df.DATETIME.dt.date)["TOTAL COUNTs"].sum()

Upvotes: 0

Related Questions