Repository navigation
BUG: cannot reshape array of size 1252 into shape (13,96) #14
Description
Activity
It seems that the problem is inside of CGMS2DayByDay implementation, due to np.interp() inability to deal with DST crossover.
We are losing/adding 1h measurements and interp_data.reshape() is unable to reshape daily measurements accordingly.The simplest solution is to convert all times to timeshift defined by the first timestamp.
- linked a pull request that will close this issueAdd Pandas 1.5.x Compatibility for Databricks Environments #15
on Sep 24, 2025
Exception observed for some rare timeseries:
cannot reshape array of size 1252 into shape (13,96)
cannot reshape array of size 3444 into shape (11,288)
cannot reshape array of size 92 into shape (0,96)