You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
DatetimeIndex with timezone information (e.g. DatetimeIndex(['2018-10-26 13:00:00-05:30'], dtype='datetime64[ns, pytz.FixedOffset(-330]',…)) are processed in xarray with numpy engine which is not timezone aware.
The index is then casted as object and it creates issues when plotting because the time scale is not considered as correct by pyplot.
The issue is being tracked by xarray there:Resample not working when time coordinate is timezone aware · Issue #1490 · pydata/xarray , but it does not seem to be moving a lot.
Maybe removing the timezone information as part of a quickfix and reintroducing it when xarray solved it?
The text was updated successfully, but these errors were encountered:
We seem to have issues with timezone aware time series, due to: pydata/xarray#1490
Would be nice to see if we can do something about it.
Description written by @Bugdup
DatetimeIndex with timezone information (e.g.
DatetimeIndex(['2018-10-26 13:00:00-05:30'], dtype='datetime64[ns, pytz.FixedOffset(-330]',…)
) are processed in xarray withnumpy
engine which is not timezone aware.The index is then casted as object and it creates issues when plotting because the time scale is not considered as correct by pyplot.
The issue is being tracked by xarray there:Resample not working when time coordinate is timezone aware · Issue #1490 · pydata/xarray , but it does not seem to be moving a lot.
Maybe removing the timezone information as part of a quickfix and reintroducing it when xarray solved it?
The text was updated successfully, but these errors were encountered: