xs.climatological_mean and xs.compue_deltas unstack data into [year, month, day] to perform their actions, then puts everything back into a 1D time array using:
ds = ds.stack(time=("year", "month", "day")) (line 123 in climatological_mean)
This might have been an oversight on our part, but this only works when all values of day exist for every month. With daily data, this creates an array of size 372 (31 x 12), including many impossible days that make pandas crash.
Steps To Reproduce
No response
Additional context
No response
Contribution
[ ] I would be willing/able to open a Pull Request to address this bug.
Setup Information
Description
xs.climatological_meanandxs.compue_deltasunstack data into[year, month, day]to perform their actions, then puts everything back into a 1Dtimearray using:ds = ds.stack(time=("year", "month", "day"))(line 123 in climatological_mean)This might have been an oversight on our part, but this only works when all values of
dayexist for every month. With daily data, this creates an array of size 372 (31 x 12), including many impossible days that makepandascrash.Steps To Reproduce
No response
Additional context
No response
Contribution