Root directory containing space_tx formatted experiment

if set, uses GPU instead of CPU

Expected diameter of cells. Should be 0 if a custom model is used.

If true and decoded_loc is provided, mrna data will be used in calculations.

The views to use for cellpose segmentation.

Location of directory that is output from the starfishRunner step, only needed if mRNA information is to be included.

The number of z-planes in each image. All that matters is whether this is equal to 1 or not, retaining the same var name as conversion for simplification.

If not None, removes cytoplasms whose nuclei lie within the given distance from the border.

If provided, segmentation will only be run on FOVs with these indices.

threshold for filtering cell segmentations (increasing this will filter out lower confidence segmentations), range is 0 to infinity

Pixel size labels are dilated by in final step. Helpful for closing small holes that are common from thresholding but can also cause cell boundaries to exceed their true boundaries if set too high. Label dilation respects label borders and does not mix labels.

json containing step parameters.

maximum size for a cell (in pixels)

minimum size for a cell (in pixels)

threshold for stitching together segmentations that occur at the same xy location but in adjacent z slices, range is 0 to 1. This should only be used when the image is 3D.

determines the extent of the segmentations (0 is the default more negative values result in larger cells, more positive values result in smaller cells), range is -6 to 6.

Manually trained cellpose model to use.

Cellpose-provided model to use.