search

seung-lab / igneous

Scalable Neuroglancer compatible Downsampling, Meshing, Skeletonizing, Contrast Normalization, Transfers and more.
GNU General Public License v3.0
40 stars 16 forks source link

Some volumes are not Downsampled #157

Closed copypasteearth closed 5 months ago

copypasteearth commented 10 months ago

Hello,

I am trying to downsample some cloud volumes, and when i convert images to volumes, they will downsample, but then I am doing some chunck wise processing and the resulting volume will not downsample

WARNING: No scales generated.
Volume Bounds:  Bbox([0, 0, 0],[1296, 1296, 26], dtype=int32)
Selected ROI:   Bbox([0, 0, 0],[1296, 1296, 26], dtype=int32)
Tasks:   0%|          | 0/4 [00:00<?, ?it/s]No factors generated. Image Shape: (272, 1024, 26, 1), Downsample Shape: [1024 1024   26], Volume Shape: [1296 1296   26], Bounds: Bbox([1024, 0, 0],[1296, 1024, 26], dtype=int32)
No factors generated. Image Shape: (1024, 1024, 26, 1), Downsample Shape: [1024 1024   26], Volume Shape: [1296 1296   26], Bounds: Bbox([0, 0, 0],[1024, 1024, 26], dtype=int32)
Tasks:  25%|██▌       | 1/4 [00:00<00:00,  5.37it/s]No factors generated. Image Shape: (1024, 272, 26, 1), Downsample Shape: [1024 1024   26], Volume Shape: [1296 1296   26], Bounds: Bbox([0, 1024, 0],[1024, 1296, 26], dtype=int32)
No factors generated. Image Shape: (272, 272, 26, 1), Downsample Shape: [1024 1024   26], Volume Shape: [1296 1296   26], Bounds: Bbox([1024, 1024, 0],[1296, 1296, 26], dtype=int32)
Done!

Can you give me an idea why these will not downsample, is it because of the 1 in the shape?

william-silversmith commented 10 months ago

Looks like the volume size is pretty small. What chunk size are you using? Setting a smaller size will probably fix it.

On Sun, Aug 13, 2023, 1:26 PM John Rowan @.***> wrote:

Hello,

I am trying to downsample some cloud volumes, and when i convert images to volumes, they will downsample, but then I am doing some chunck wise processing and the resulting volume will not downsample

WARNING: No scales generated. Volume Bounds: Bbox([0, 0, 0],[1296, 1296, 26], dtype=int32) Selected ROI: Bbox([0, 0, 0],[1296, 1296, 26], dtype=int32) Tasks: 0%| | 0/4 [00:00<?, ?it/s]No factors generated. Image Shape: (272, 1024, 26, 1), Downsample Shape: [1024 1024 26], Volume Shape: [1296 1296 26], Bounds: Bbox([1024, 0, 0],[1296, 1024, 26], dtype=int32) No factors generated. Image Shape: (1024, 1024, 26, 1), Downsample Shape: [1024 1024 26], Volume Shape: [1296 1296 26], Bounds: Bbox([0, 0, 0],[1024, 1024, 26], dtype=int32) Tasks: 25%|██▌ | 1/4 [00:00<00:00, 5.37it/s]No factors generated. Image Shape: (1024, 272, 26, 1), Downsample Shape: [1024 1024 26], Volume Shape: [1296 1296 26], Bounds: Bbox([0, 1024, 0],[1024, 1296, 26], dtype=int32) No factors generated. Image Shape: (272, 272, 26, 1), Downsample Shape: [1024 1024 26], Volume Shape: [1296 1296 26], Bounds: Bbox([1024, 1024, 0],[1296, 1296, 26], dtype=int32) Done!

Can you give me an idea why these will not downsample, is it because of the 1 in the shape?

— Reply to this email directly, view it on GitHub https://github.com/seung-lab/igneous/issues/157, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATGQSPDBP7VFQEG36VVGE3XVEE3BANCNFSM6AAAAAA3OYZ6ZQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

copypasteearth commented 10 months ago

@william-silversmith thank you, lowering the chunk size fixed it.