openclimatefix / openclimatefix/pvnet
Image augmentation - noise added
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 60
- Forks
- 50
- Avg merge
- 1d 21h
- Merged PRs (30d)
- 2
Description
Relating directly to ML Ideas 12:
Goal to enhance robustness and generalisation of PVNet. To implement a set of specific image augmentation techniques, applied purely to the training data as a fundamental extra process stage.
Augmentation types to be considered such as small, random rotations - affine transformations. Additive Gaussian noise also certainly. Minimised rotations / stretching etc for NWP data.
Initial proposal to use the Albumentations library. Augmentations applied within the getitem method of the Dataset classes that handle satellite and NWP data - logic situated in the utils file itself.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Locate the Dataset classes and their getitem methods, along with the relevant utils file, and review how satellite and NWP training data currently enter the pipeline. Read the Albumentations proposal and define the augmentation set and constraints for rotations, affine changes, and Gaussian noise. Done means the agreed augmentations run only on training data for both data sources and are covered by suitable checks.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- data, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 28/100