[main] Add spatialdata element to segger export - #80

Open

Tobiaspk wants to merge 18 commits into

mainfrom

feature/spatialdata_export

Commits on Aug 31, 2026

Initial spatialdata interface

tobiaspk
authored and
Tobiaspk
committed
95786d1
Use new 'filtered' column and remove faulty min_similarity validators

tobiaspk
authored and
Tobiaspk
committed
3fb3775
Make -i/source_path optional; fall back to joining source transcripts for legacy segmentation outputs

tobiaspk
authored and
Tobiaspk
committed
29c2011
Add example usage to export docstring

tobiaspk
authored and
Tobiaspk
committed
4ea4f50
Log copy/write start for spatialdata export; check destination existence upfront

tobiaspk
authored and
Tobiaspk
committed
5e71342
Check spatialdata package is importable before exporting the 'spatialdata' element

tobiaspk
authored and
Tobiaspk
committed
785fd35
add spatialdata dependency

tobiaspk
authored and
Tobiaspk
committed
35b7acf
Parallelize cell boundary generation across CPUs

tobiaspk
authored and
Tobiaspk
committed
18bcc50
Add setuptools dependency to cuda121 env
Tobiaspk
committed
dc9c7d5
Fix parallel boundary generation: list unpacking and worker count
Tobiaspk
committed
d48625a
Write spatialdata export elements in place, avoid dask-expr crash
Tobiaspk
committed
d1ec959
Thread an optional z coordinate through export and training writers

committed
b0e198c
Append segger columns to existing transcripts element instead of a new one
Tobiaspk
committed
6515b6e
Simplify export.py to rely on the 'filtered' column directly
Tobiaspk
committed
a1e223b
Take element names as separate args in _check_sdata_elements
Tobiaspk
committed
7ec59c1
Propagate base-element transformations in spatialdata export, merge lazily in dask
Tobiaspk
committed
047716e

Navigation Menu

Uh oh!

[main] Add spatialdata element to segger export - #80

[main] Add spatialdata element to segger export#80
Tobiaspk wants to merge 18 commits into
mainfrom
feature/spatialdata_export

Commits on Aug 31, 2026

Initial spatialdata interface

Use new 'filtered' column and remove faulty min_similarity validators

Make -i/source_path optional; fall back to joining source transcripts for legacy segmentation outputs

Add example usage to export docstring

Log copy/write start for spatialdata export; check destination existence upfront

Check spatialdata package is importable before exporting the 'spatialdata' element

add spatialdata dependency

Parallelize cell boundary generation across CPUs

Add setuptools dependency to cuda121 env

Fix parallel boundary generation: list unpacking and worker count

Write spatialdata export elements in place, avoid dask-expr crash

Thread an optional z coordinate through export and training writers

Append segger columns to existing transcripts element instead of a new one

Simplify export.py to rely on the 'filtered' column directly

Take element names as separate args in _check_sdata_elements

Propagate base-element transformations in spatialdata export, merge lazily in dask

Commits on Sep 1, 2026

Fix in-place spatialdata overwrite and drop row_index from transcripts

Add --min-counts filter and align export table to boundaries