CliMA / CliMA/EnsembleKalmanProcesses.jl
O3.7.5 Formulate and implement variant of EKI that scales linearly in output dimension and allows for sampling error correction
- Dominant language
- Julia
- Stars
- 125
- Forks
- 24
- Avg merge
- 1d 18h
- Merged PRs (30d)
- 5
Description
# Purpose
To implement some changes that (1) make EKI (scalable in input dim) more scalable in output dimensions, and (2) make ETKI (scalable in output dim) more scalable in input dim
```[tasklist]
- [ ] https://github.com/CliMA/EnsembleKalmanProcesses.jl/issues/377
- [ ] EKI update linear algebra from online->offline to reduce online complexity
- [ ] Implement Localization implementations that is compatible with ETKI
```
Contributor guide
No contributing guide indexed for this repository
Research direction
Start with linked issue #377 and the existing EKI and ETKI implementations; this issue does not identify files, tests, or entry points. Review the requested online-to-offline linear algebra and ETKI-compatible localization work. Done means the requested variants support the stated output and input scaling goals and sampling error correction.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100