this is more a suggestion than an issue.
I think it would be more efficient to allow the reducer to work directly on the stored .nc files in the clusters. Instaed of downloading the correct file and interpret them, why cannot the reducer directly reduce the correct file and interpret them?
in any case thank you for this very useful package
Iv
this is more a suggestion than an issue.
I think it would be more efficient to allow the reducer to work directly on the stored .nc files in the clusters. Instaed of downloading the correct file and interpret them, why cannot the reducer directly reduce the correct file and interpret them?
in any case thank you for this very useful package
Iv