Skip to content

Releases: pattonw/mwatershed

Add randomized_strides

09 Oct 17:20
Compare
Choose a tag to compare

When calling mwatershed.agglom you can now pass randomized_strides=True. For a given offset/stride pair, the probability of picking an edge is equal to 1/prod(stride).

v0.5.1

15 Mar 00:16
Compare
Choose a tag to compare

Fix bug in agglomerate function when stride was set to None. In this case we generate default strides of 1x1x... however instead of generating one stride per offset we were generating one stride per dimension, resulting in many offsets being ignored in applications using long range affinities

Add strides

06 Mar 17:07
Compare
Choose a tag to compare

Arbitrary strides are now supported for each offset in your neighborhood.

v0.4

29 Dec 19:09
Compare
Choose a tag to compare

Fix bug in the processing of user defined edges

v0.3

29 Dec 18:11
Compare
Choose a tag to compare

Filter out the background.

Given an array of affinities, voxels will be merged and given ids greater than 0. If a voxel has no positive edges it is considered part of the background and will be assigned 0 unless it was given a seed label by the user, in which case the label will not change.

v0.2

31 May 16:43
Compare
Choose a tag to compare

add support for negative offsets

v0.1

11 May 22:14
Compare
Choose a tag to compare

initial release