option to broadcast across multiple exposures #40
Description
If taking multiple exposures/point, the dark plan currently must also take that number of exposures. In other words, the dark is only broadcast across dim0 of array. it would be better to have options to
a.) take multiple exposures as it does now
b.) take one exposure (broadcast across both dim0 and dim 1) - to do this, the dark plan needs to change the n exposures, take the dark, and then change it back.
c.) take multiple exposures but subtract the mean of the multiple exposures rather than each respective exposure.
Scientifically speaking b and c seem much both much more relevant, with b saving a lot of time, and c resulting in a reduction in noise. There might be a case for a in some cases if the detector has some repeatable dynamic response (e.g. heating up) while taking multiple exposures.
Perhaps a clean option is to take whatever dimensionality that the dark plan produces. if the dark plan only produces one frame, then broadcast it. If it produces m != n frames, average and subtract the average from all frames, if it produces n frames, subtract them from the respective frame.