Skip to content

Development

This page tracks current limitations, implementation status, and roadmap items.

Known Issues

  • Outputs can still look somewhat speckled/noisy.
  • Patches with large land coverage can degrade generation quality across the full patch.

Potential mitigation directions already identified: - DDIM sampling for better compute/fidelity tradeoff - structure-aware or frequency-aware losses - parameterization/schedule tuning

Implementation Status Notes

  • mask_loss_with_valid_pixels: implemented and working
  • coord_conditioning: implemented and tested
  • lightweight dataset + datamodule path: implemented and working
  • x0 parameterization: implemented and working well in current experiments
  • combined date + coordinate embedding: implemented, now exercised in EO config

ToDos

  • [ ] DDIM sampling path still needs deeper validation across checkpoints/settings
  • [ ] Increase U-Net capacity (for example dim: 64 -> 96/128, deeper dim_mults)
  • [ ] Add frequency-aware objectives (for example gradient/PSD losses) to reduce speckle noise
  • [ ] Activate and validate EMA weights in full training runs

Done

  • [x] Encode timestamps together with coordinate embedding
  • [x] Add and test x0 parameterization path
  • [x] Establish geographically consistent window split tooling
  • [x] Implement known-pixel clamping mechanism for sampling
  • [x] Use larger corruption patches instead of isolated single pixels
  • [x] Add dataset-to-disk export pipeline
  • [x] Implement masked loss support for land/validity handling
  • [x] Maintain dependency list in repository

Roadmap

Tier 1

  • [x] Aux priors via patch-level FiLM conditioning from coordinates (and optional date)
  • [ ] Increase sparse-input stress test to mask_fraction=0.99 as a standard comparison setting
  • [ ] Implement trajectory-style corruption ("walk" masks) to better simulate submarine-like movement across each patch
  • [ ] Simulate EO observation + sparse in-situ depth profile setup more systematically
  • [ ] Evaluate lower-resolution setups aligned with expected Argo-like profile density

Tier 2

  • [ ] Evaluate additional Copernicus Marine products (for example ARMOR3D)
  • [x] Improve mask handling design in conditional inputs
  • [ ] Explore stronger backbones