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 iohub Position.append_channel() function works great when the array to have a channel appended to shares the same dtype. In this case, we have some datasets stored as dtype=np.uint16 and when attempting to append a np.float32 dtype array, this proceeds without a warning.
The appropriate behaviour should be to raise a warning. In the case of having to change the whole dtype, I think it is probably easier to create a new zarr store, copy in the data, and then delete the old zarr store.
The text was updated successfully, but these errors were encountered:
iohub.ngff.Position.append_channel only takes in the new channel name and a flag to resize arrays. It does not take any data as input. Where do you think the warning should be triggered?
The iohub
Position.append_channel()
function works great when the array to have a channel appended to shares the samedtype
. In this case, we have some datasets stored asdtype=np.uint16
and when attempting to append anp.float32
dtype array, this proceeds without a warning.The appropriate behaviour should be to raise a warning. In the case of having to change the whole dtype, I think it is probably easier to create a new zarr store, copy in the data, and then delete the old zarr store.
The text was updated successfully, but these errors were encountered: