[CharlelieLrt]

Description

This PR adds TrigFlowNoiseScheduler and TrigFlowPreconditioner to
physicsnemo.diffusion. Users can pair them with the same sigma_data value to train
a backbone that predicts normalized velocity while exposing clean-data
predictions through the usual diffusion model interface.

Built-in linear Gaussian schedules also gain snr_inv(q), which converts the
amplitude ratio q = alpha / sigma back to diffusion time.
Custom schedules can supply their own inverse. This gives the private FastGen
adapters a way to translate noise levels between schedules without relying on
inference grids.

Checklist

• This PR follows the contribution guidelines.
• The documentation describes the API and its role in the stack.
• This PR remains a draft.

Dependencies

• Depends on

  • PhysicsNeMo #2094 adds named diffusion inputs and outputs. This is the immediate parent branch.

• Depends on this PR

  • PhysicsNeMo #2096 adds private adapters and public sCM and MeanFlow losses. It follows this PR directly in the stack.

• Later follow-ups

  • PhysicsNeMo #2097 extends the loss API to DMD2 and f-distill.
  • PhysicsNeMo #2098 adds checkpoint bundles for restarting distillation.
  • PhysicsNeMo #2099 adds the optional training strategy and the grouped changelog entry.

• Parallel FastGen work

  • FastGen #43 allows the NumPy 2 dependency used by PhysicsNeMo. PhysicsNeMo Add sCM and MeanFlow distillation losses #2096 uses an immutable fork pin, so it does not need to wait for this proposal to merge.
  • FastGen #44 provides shared loss hooks for external training loops. The initial PhysicsNeMo integration does not require these hooks.

• Conditional follow-up

  • PhysicsNeMo #2100 replaces local compatibility code with the proposed FastGen loss hooks. It depends on PhysicsNeMo Add optional diffusion distillation strategy #2099 and FastGen [other] Add modulus.models.fcn_mip_plugin #44, not the other way around.

If FastGen maintainers decline FastGen #44, we would close PhysicsNeMo #2100.
The preceding PhysicsNeMo stack would remain available.

Review process

Reviewers can compare this PR with its immediate parent. After a parent merges,
we update the remaining branches and their PR bases to keep each diff focused.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[github-actions]

CODEOWNERS review map

Current for commit bda6a9107929. An approval covers every file listed for that owner; one owner is sufficient for shared files.

⏳ @CharlelieLrt — 24 file(s)

• docs/api/diffusion/noise_schedulers.rst
• docs/api/diffusion/preconditioners.rst
• physicsnemo/diffusion/noise_schedulers/init.py
• physicsnemo/diffusion/noise_schedulers/edm.py
• physicsnemo/diffusion/noise_schedulers/iddpm.py
• physicsnemo/diffusion/noise_schedulers/linear_gaussian.py
• physicsnemo/diffusion/noise_schedulers/rectified_flow.py
• physicsnemo/diffusion/noise_schedulers/trigflow.py
• physicsnemo/diffusion/noise_schedulers/ve.py
• physicsnemo/diffusion/noise_schedulers/vp.py
• physicsnemo/diffusion/preconditioners/init.py
• physicsnemo/diffusion/preconditioners/preconditioners.py
• test/diffusion/data/trigflow_precond_conv.mdlus
• test/diffusion/data/trigflow_precond_conv_coefficients.pth
• test/diffusion/data/trigflow_precond_conv_forward.pth
• test/diffusion/data/trigflow_precond_conv_sigma.pth
• test/diffusion/data/trigflow_precond_linear.mdlus
• test/diffusion/data/trigflow_precond_linear_coefficients.pth
• test/diffusion/data/trigflow_precond_linear_forward.pth
• test/diffusion/data/trigflow_precond_linear_sigma.pth
• test/diffusion/test_model_io.py
• test/diffusion/test_noise_schedulers.py
• test/diffusion/test_preconditioners.py
• test/diffusion/test_trigflow.py

⏳ @megnvidia — 2 file(s)

• docs/api/diffusion/noise_schedulers.rst
• docs/api/diffusion/preconditioners.rst

No CODEOWNER

• CHANGELOG.md

Comment /codeowners-info to refresh.

[CharlelieLrt]

Continued in PhysicsNeMo #2200, from branch diffusion-trigflow in CharlelieLrt/modulus. The replacement preserves this implementation and includes the requested documentation clarifications. Please continue the review there. This PR retains the earlier discussion.