Compute mean for multi group of panda dataframe rows-CodePudding

From a panda dataframe with theses columns:

DAT_MESURE         datetime64[ns]
MES_TEMPERATURE            object

and values:

I want to compute the mean temperature value for a group of hours and get a new df. For example, I want to create a new df with DAT_MESURE rounded to lowest hour and a mean for 4 values of the hour.

I want to get:

DAT_MESURE MES_TEMPERATURES
2020-08-01 00:00:00 21,xx
2020-08-01 01:00:00 22,xx

How to code it in python panda please?

CodePudding user response：

Use:

df['MES_TEMPERATURE'] = df['MES_TEMPERATURE'].str.replace(',','.', regex=True).astype(float)

df1 = df.resample('H', on='DAT_MESURE')['MES_TEMPERATURE'].mean()

Or:

df2 = df.groupby(df['DAT_MESURE'].dt.floor('H'))['MES_TEMPERATURE'].mean()

If need round:

df3 = df.groupby(df['DAT_MESURE'].dt.round('H'))['MES_TEMPERATURE'].mean()