zouinkhim

2D students to top-level setup_53_2d/ setup_54_2d/ (3D manifest schema, per-axis sizes, step 20000); source_repo on 3D manifests; remove 2d/

2c17b0b verified 7 days ago

preview code

Raw History Blame Contribute Delete

712 Bytes

# setup_54_2d — 2D-slice mito student (experimental)

One z-slice in, one slice out: `input` float32 `[1,1,1,N,N]` → `output` float32 raw logits `[1,3,1,N-88,N-88]`
(mito affinities z, y, x; threshold at 0). 16 nm, input norm `uint8/255*2-1`.

Distilled from the 3D model `setup_54` (PyTorch: `cellmap/mito-aff-resunet-setup-54`): every 3D kernel collapsed to 2D (sum of depth taps) and
fine-tuned to reproduce the 3D predictions slice by slice. Training step, per-file fidelity and sizes are in
`manifest.json` (same schema as the 3D folders, per-axis `input_size` / `output_size`).

cellmap-flow serverless: Input size `1,N,N`, Output size `1,N-88,N-88`, voxel 16,16,16, Channels 3, Has channel axis true.

	# setup_54_2d — 2D-slice mito student (experimental)

	One z-slice in, one slice out: `input` float32 `[1,1,1,N,N]` → `output` float32 raw logits `[1,3,1,N-88,N-88]`
	(mito affinities z, y, x; threshold at 0). 16 nm, input norm `uint8/255*2-1`.

	Distilled from the 3D model `setup_54` (PyTorch: `cellmap/mito-aff-resunet-setup-54`): every 3D kernel collapsed to 2D (sum of depth taps) and
	fine-tuned to reproduce the 3D predictions slice by slice. Training step, per-file fidelity and sizes are in
	`manifest.json` (same schema as the 3D folders, per-axis `input_size` / `output_size`).

	cellmap-flow serverless: Input size `1,N,N`, Output size `1,N-88,N-88`, voxel 16,16,16, Channels 3, Has channel axis true.