Add overlap-aware vessel skeleton graph export (skeleton + SDT → sparse parquet)#167
Conversation
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/81c57d7a-d4d2-4e76-8d37-74454b1d2bbe Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/81c57d7a-d4d2-4e76-8d37-74454b1d2bbe Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/81c57d7a-d4d2-4e76-8d37-74454b1d2bbe Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/81c57d7a-d4d2-4e76-8d37-74454b1d2bbe Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
|
@copilot: encountering an error when building the graph. I've fixed the dependencies and added a small vessel mask and skeleton mask (in tests/example_vessels_mask.nii.gz and tests/example_skeleton.nii.gz), make some test cases to use those to iron out these errors. these examples come from a cropped region of a real ome zarr dataset, if loaded with zarrnii into a chunked dask array they can act as appropriate test datasets (smaller chunks and smaller overlaps may need to be used for this test case, since these are 300x300x300.. Traceback (most recent call last): |
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/28093ce7-f351-495d-8ffb-e07a04bc9f2e Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/28093ce7-f351-495d-8ffb-e07a04bc9f2e Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/28093ce7-f351-495d-8ffb-e07a04bc9f2e Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/28093ce7-f351-495d-8ffb-e07a04bc9f2e Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Implemented in commit |
|
@copilot: still having same error. Also I had forgotten to push the example nii.gz images those are there now see if you can reproduce the error and use those images for testing. Traceback (most recent call last): |
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Agent-Logs-Url: https://github.com/khanlab/SPIMquant/sessions/44ded292-ec14-4f34-b408-58969580e084 Co-authored-by: akhanf <11492701+akhanf@users.noreply.github.com>
Reproduced and fixed in |
|
works now, will do another PR to produce separate nodes and edges tables |
Uh oh!
There was an error while loading. Please reload this page.