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
There are many signals where the dtype is known a-priori to be floats. However, Ophyd always looks at an example value when describe is called, and shoves that into a data_type function.
If the first value happens to be an integer, the describe method then returns integer, and all further data will be cast to an integer, even if everything else is floats. I think the true data is saved, somehow, but have never really managed to get it back out of Tiled.
There should be a way to tell a signal to treat numeric dtypes as number and not integer, so that this can be set properly on a signal-by-signal basis.
The text was updated successfully, but these errors were encountered:
Summary
There are many signals where the dtype is known a-priori to be floats. However, Ophyd always looks at an example value when
describe
is called, and shoves that into adata_type
function.If the first value happens to be an integer, the
describe
method then returns integer, and all further data will be cast to an integer, even if everything else is floats. I think the true data is saved, somehow, but have never really managed to get it back out of Tiled.There should be a way to tell a signal to treat numeric dtypes as
number
and notinteger
, so that this can be set properly on a signal-by-signal basis.The text was updated successfully, but these errors were encountered: