Greg M. Fleishman

Distributed cellpose using Dask for large (3D) images

3

Hi Cellpose Team! This is quite late, but per discussions here: https://github.com/MouseLand/cellpose/issues/244 This is the distributed Cellpose that I am using for large 3D volumes (~2K x 6K x 5K...

Hi Paul, Was really hoping to register using my full resolution data, but there seems to be a scaling issue. ITK version: 4.13.2 Greedy commit: abef6386366cd1a7d296cd8df0d76c9b3368aaab I'm using an old...

contrib: Alternative distributed_segmentation, automated merge/split tools, ellipsoid shape filter

6

All contributions are in their own modules in the contrib folder - no cellpose files were modified. There is an existing distributed_segmentation in contrib well written by the knowledgeable @chrisroat....

Getting in touch

1

Hi Marvin, I didn't know a better way to contact you, but I'd love to talk more about this registration code. I have built some very similar tools using dask...

NMS implementation does not scale

9

I've proposed a solution to this issue in a pull request #40 . The [NMS implementation](https://github.com/mpicbg-csbd/stardist/blob/f8e63dc3fe1c662b4ee71b2cc54d01728b8bd6a5/stardist/lib/stardist3d_impl.cpp#L929-L1306) scales with the square of the number of candidate polygons, i.e. for N candidate...

Dask task graphs are huge (non-performant) for large data

19

I think this will be a tougher issue to solve than typos, math bugs, and formatting; so it will warrant some conversation with @thewtex and @tbirdso. I have `ngff-zarr` running...

np.bool8 is deprecated since numpy v 1.24: https://github.com/numpy/numpy/issues/23755 This line should be modified: https://github.com/ramonhagenaars/nptyping/blob/785cd07e65f992f47256398fd01f62067928d29c/nptyping/typing_.py#L51