1. In: df.head()
  3. out:
  5. date time MwH
  6. 1.01.2018 00:00 27412
  7. 1.01.2018 01:00 26324
  8. 1.01.2018 02:00 24635
  9. 1.01.2018 03:00 23872
  10. 1.01.2018 04:00 23194
  11. In:
  12. df.time = pd.to_timedelta(df.time + ":00", unit="h")
  14. df['date'] = pd.to_datetime(df.date)
  16. df['date1'] = pd.to_datetime(df['date'] + df ['time'])
  18. df = df.drop(['date', 'time'], axis=1)
  20. df.info()
  23. Out:
  24. <class 'pandas.core.frame.DataFrame'>
  25. RangeIndex: 17544 entries, 0 to 17543
  26. Data columns (total 2 columns):
  27. # Column Non-Null Count Dtype
  28. --- ------ -------------- -----
  29. 0 MwH 17544 non-null int64
  30. 1 date1 17544 non-null datetime64[ns]
  31. dtypes: datetime64[ns](1), int64(1)
  32. memory usage: 274.2 KB
  34. In: df.set_index("date1", inplace = True)
  35. df = df.asfreq('h')
  36. df.index.freq = "H"
  38. train, test = df.iloc[:10000, 0], df.iloc[10500:, 0]
  40. model = est(train, trend="add", damped=False, seasonal="add", seasonal_periods=24).fit()
  42. tah = model.predict(start = test.index)
  43. '''