0

How do I refer to extracted start and end dates in order to sample a timeseries with date_range?

I get now 'DataFrame' object has no attribute 'date_range'.' (I would be happy with any solution that does not cause error with the sampling start and end dates that differ typically for different SNs). 

In: 

import pandas as pd
import numpy as np

df = pd.read_csv("csv.csv", sep=';', error_bad_lines=False) 
df['TimeDate'] = pd.to_datetime(df['TimeDate'],dayfirst=True) 

gdf = df.groupby(['SN'],as_index = False)
dates = pd.DataFrame(gdf.nth(0)) # pd.DataFrame used here to avoid a SettingwithCopyWarning.
dates['StartD'] = gdf.nth(0)['TimeDate'].array
dates['EndD'] = gdf.nth(-1)['TimeDate'].array

# Little bit odd -> managed to write the first rows only.
df = df.join(dates.reindex(columns=['StartD', 'EndD']))

# Grouping by 'SN'. -> Trying to samnple
gdf = df.groupby(['SN'],as_index = False)
for k, gp in gdf:
             print('key=' + str(k))
             print(gp.date_range(start=gdf['StartD'], end=gdf['EndD'], freq='D'))

Out:
       SN         TimeDate  ...             StartD               EndD
0      sn000066   2016-09-28 00:01:00  ...  2016-09-28 00:01:00  2016-10-03 23:59:00
1      sn000066   2016-09-29 00:01:00  ... 
...  
n      sn000066   2016-10-02 00:01:00  ...
MattiH
  • 35
  • 6
  • This is how I resolved the grouping and resampling of timeseries: https://stackoverflow.com/questions/32012012/pandas-resample-timeseries-with-groupby I have no answr to How to re a datetime in a stored location – MattiH Dec 27 '19 at 19:59
  • Above reads => I have no (do not need anymore) an answer how to refer to a datetime in a stored location and pass it as an argument to groupby. I suppose this is one of the reasons pd.Grouper was developed... – MattiH Dec 27 '19 at 20:08

0 Answers0