[Lane 1 · numerics] GroupNormalization — PyTorch reference values for GroupNorm #1380

Sub-issue of #1379 ([Feature]: GroupNormalization forward pass (groupNorm {} DSL currently crashes)).

Lane: 1 · Numerics/Research
Skill needed: PyTorch. No Kotlin needed.
Size: xs (under 1 hour)
Blocked by: nothing

What to do

1. Compute torch.nn.GroupNorm outputs for a seeded [2, 6, 3, 3] input with num_groups 1, 2, 3 and 6, affine=True with non-trivial gamma/beta, and once with affine=False.
2. Post the script and the inputs/outputs (flattened lists are fine) as a comment here.
3. State whether PyTorch uses the biased or unbiased variance (and link the doc that says so).

Acceptance

• Script + values posted
• Variance convention stated with a source
• Result reported back on the parent issue

Notes

If something here turns out to be wrong or much bigger than the size label, comment on this issue or open a specific bug instead of blocking the parent.

Coverage matrix: docs/modules/ROOT/pages/reference/ops-status-matrix.adoc · How lanes work: Issue taxonomy