apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-11 21:12:38 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update parameter tying investigation with GitHub issue findings

Browse source
This commit is contained in:
Neil Lawrence 2025-08-15 11:16:43 +02:00
parent e659649851
commit f058078017
1 changed files with 13 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

14
backlog/infrastructure/2025-08-15_parameter-tying-framework.md
View file

@ -31,7 +31,7 @@ During LFM kernel code review, we identified that GPy lacks systematic parameter
## Investigation Needed
### 1. Scope Assessment
- [ ] Search existing GitHub issues for parameter tying discussions
- [x] Search existing GitHub issues for parameter tying discussions
- [ ] Identify other kernels/models that could benefit from parameter tying
- [ ] Assess impact on current GPy codebase
@ -74,3 +74,15 @@ During LFM kernel code review, we identified that GPy lacks systematic parameter
### 2025-08-15
Task created after identifying parameter tying as a potential limitation during LFM kernel code review. Need to investigate scope and create CIP for community discussion.
### 2025-08-15 (GitHub Investigation)
Found existing GitHub issues confirming parameter tying limitations:
- **Issue #462 (2016)**: "tie_params doesnt work ?" - `AttributeError: 'Add' object has no attribute 'tie_params'`
- **Issue #789 (2019)**: "Non-implemented Param tying work-around options" - Confirms `tie_to` from Parametrized is not implemented
- **Issue #878 (2020)**: "Constraining hyperparameters" - Open issue requesting parameter equality constraints in MultioutputGP
**Key Findings:**
- Parameter tying functionality has been missing/broken in GPy for at least 5 years
- Multiple users have requested this feature for different use cases
- Current workarounds involve manual parameter management
- No systematic solution exists in the codebase