I got the following dataframe as an example using pandas:
| Timestamp | Computer no. | Memory Usage in % |
|---|---|---|
| 05-07-2022 08:00:00 | 1 | 25 |
| 05-07-2022 08:00:00 | 2 | 56 |
| 05-07-2022 08:00:00 | 3 | 34 |
| 05-07-2022 08:15:00 | 1 | 23 |
| 05-07-2022 08:15:00 | 2 | 63 |
| 05-07-2022 08:15:00 | 3 | 23 |
| 05-07-2022 08:30:00 | 1 | 26 |
| 05-07-2022 08:30:00 | 2 | 56 |
| 05-07-2022 08:30:00 | 3 | 32 |
| 05-07-2022 08:45:00 | 1 | 23 |
| 05-07-2022 08:45:00 | 2 | 15 |
| 05-07-2022 08:45:00 | 3 | 86 |
| 05-07-2022 09:00:00 | 1 | 43 |
| 05-07-2022 09:00:00 | 2 | 26 |
| 05-07-2022 09:00:00 | 3 | 21 |
I would like to aggregate the time from a 15 Minute intervall to 1 hour. But I have different Computer No. in the same dataset. Firstly how do I aggregate the time to 1 hour and secondly how do I do it for each computer seperatly?
Thanks!
CodePudding user response：
You can extract the hour value from the timestamp and then groupby -
df['hour'] = pd.to_datetime(df['Timestamp']).dt.hour
df.groupby(['hour', 'Computer no.']).agg('mean').reset_index()
Output
hour Computer no. Memory Usage in %
0 8 1 24.25
1 8 2 47.50
2 8 3 43.75
3 9 1 43.00
4 9 2 26.00
CodePudding user response：
This should work:
df = df.set_index('Timestamp')
df1 = df[df['Computer no.']==1].resample('1H').mean()
df2 = df[df['Computer no.']==2].resample('1H').mean()
df3 = df[df['Computer no.']==3].resample('1H').mean()