Track IDs of random effects that fail to converge
Open
@ashelkovnykov is already working on this.
Since Apr 11, 2017.
enhancement
- Dominant language
- Terra
- Stars
- 795
- Forks
- 175
- PR merge metrics
- No merged PRs in 30d
Description
When optimization tracking is enabled, random effects currently track and summarize the convergence reason (or lack thereof) for each ID. The IDs that failed to converge should be logged and output at the end of training to aid tracking bugs.
It's also important to question: what causes models to fail to converge - can we detect it before training through data validation alone?
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.
Assessment
This issue has not been assessed yet.