CODEWITHSUNDEEP
pythonpandastime-series
Sashaank
Sashaank

Reputation: 964

How to resample uneven time stamps per minute

I am working on some data where data is recorded every few seconds for an entire day. I have to resample this data by taking the mean for every minute.

For example here is my data

Date    Time    Recorded_data
2019-01-01    00:00:20    0.20
2019-01-01    00:00:25    0.25
2019-01-01    00:00:28    0.16
2019-01-01    00:00:32    0.26
2019-01-01    00:00:36    0.28
2019-01-01    00:00:45    0.26
2019-01-01    00:00:48    0.24
2019-01-01    00:00:56    0.24
2019-01-01    00:01:00    0.18
...

I want to resample the data above as

Date    Time    Recorded_data
2019-01-01    00:00:00    0.23
...

I want to take the mean for all data for every minute.

I tried resample with pandas but that just converts all data in the Recorded_data column to Nan

this is the code i used to resample the data

df.Recorded_data.resample('min').mean()

I tried the answer in this link, but it does not seem to work

Upvotes: 0

Views: 251

Answers (2)

jezrael
jezrael

Reputation: 862641

Use DataFrame.resample by T for minutes with mean with parameter on, then DatetimeIndex is not necesary:

df['datetime'] = pd.to_datetime(df['Date'] + ' ' + df['Time'])

df = df.resample('T', on='datetime').Recorded_data.mean().reset_index(name='Recorded_data')
print (df)
             datetime  Recorded_data
0 2019-01-01 00:00:00        0.23625
1 2019-01-01 00:01:00        0.18000

Upvotes: 1

Vaishali
Vaishali

Reputation: 38415

If Date and Time are separate columns, combine them first and convert to date time. You can then resample on the datetime column,

df['datetime'] = pd.to_datetime(df['Date'] + ' ' + df['Time'])

df.set_index('datetime').resample('H').Recorded_data.mean().reset_index(name = 'Recorded_data')

You get

    datetime    Recorded_data
0   2019-01-01  0.23

Upvotes: 2

Related Questions