Question on extracting Aurora latent states
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1k
- Forks
- 174
- PR merge metrics
- No merged PRs in 30d
Description
Hi Aurora team,
I am currently working on using aurora for weather prediction, and thank you very much for the impressive work you’ve done on the model.
We are now exploring a way on how to extract Aurora's latent state at each time step and then to have a better understanding of the predictability. For example, which latent states are most important at different times or flow regimes, and how the relative importance of these latent states evolves over time.
Is this feasible and if so is it possible to recommend some ways to obtain these latent states from Aurora?
Thank you again for your work and your time.
Contributor guide
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
The issue does not name a file, test, or entry point. Start by locating Aurora's model inference and state-handling interfaces, then determine whether latent states are exposed at each time step. Done would require a clearly scoped implementation or documented approach for extracting and inspecting those states.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100