Lost metadata when resampling

SeitaBV / timely-beliefs

Model data as beliefs (at a certain time) about events (at a certain time).

MIT License

34 stars 6 forks source link

Lost metadata when resampling #22

Closed Flix6x closed 3 years ago

Flix6x commented 4 years ago

Pandas 1.1.0 fixed a bug where resampling causes the subclassed data structure to be lost.

Another effort is undertaken to preserve metadata when using groupby, a fix for which is scheduled for release version 1.1.1.

That means by updating our dependency to pandas=1.1.1 we should be able to simplify (and increase the speed of) some of our resampling methods by not having to reinitialize the BeliefsDataFrame. I already had a test stashed for just this case.

Flix6x commented 4 years ago

Release of this fix has been postponed, at least to pandas 1.1.3.

Flix6x commented 3 years ago

Fix was released as part of pandas 1.1.5.