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
The new version of fitscube (0.5.0) allows larger than memory cubes to be created through a combination of asyncio and stashing a data cube to disk before writing to is.
Largely this should be no real issue with the exception of the use of fitscube to co-add wsclean sub-band images together. In the current version we reshuffle dimensions in the cube around for linmos comparability, and we do this using the open hdulist object the older fitscube api returns. We should drop this and simply use the approach option in the linmos parset.
The new version of fitscube (0.5.0) allows larger than memory cubes to be created through a combination of asyncio and stashing a data cube to disk before writing to is.
Largely this should be no real issue with the exception of the use of fitscube to co-add wsclean sub-band images together. In the current version we reshuffle dimensions in the cube around for linmos comparability, and we do this using the open hdulist object the older fitscube api returns. We should drop this and simply use the approach option in the linmos parset.
@AlecThomson - any other thoughts?
The text was updated successfully, but these errors were encountered: