Kevin Cutler

Ah, I eventually discovered that my code was not compiling correctly, but nevertheless a (very convincingly correct) output was being generated / displayed. It was clearly an older version of...

@VarIr that's great! Have you quantified the performance difference with the simple change (2) that you proposed?

@VarIr You are correct, I hadn't noticed that before. Not sure is there would be any benefit to it, but perhaps. Does this get in the way of the optimizations...

@athulrv, could you provide the code you used to get your results? Given the blockiness of the segmentation, I'm guessing there was some pretty dramatic rescaling of the image, and...

@athulrv, thanks for sharing that link. The scale disk is only for visualizing cell diameters, if I recall correctly. It might be useful for you to share screenshots of the...

@athulrv, thanks for sharing your data. I took a look and I think I found your problem. Your cropped image is more densely populated than the full image. If you...

@athulrv Normalize99 is not a parameter, but a function. You can correct its behavior by using the Omnipose version. You can try setting omni=True in the GUI (I think there...

Great! @carsen-stringer you may want to consider changing `normalize99` defaults in a future release.

I think I caught this bug myself in my testing and fixed it.

I was just thinking this! It would be super nice to have especially for the inversion filter on eye-searing white apps at launch.