Bug Description
When run_evals ran b2b for the same experiment, 2 ic-ops panels are active with latest not working and old ic-op panel becoming active to show new runs eventually fail to do ic-op operations
To Reproduce
Steps to reproduce the behavior:
- Run RAG-FIQA notebook, run evals b2b
- Observe the IC-OP Panels created
- Do the IC-OP operation during second run-evals is in-progress
- See error
Expected Behavior
A single ic-op panel is required to be active with ic-ops operations working
Screenshots
Environment
- OS:Ubuntu
- Python version: 3.12
- RapidFire AI version: 0.12.6
- Browser (if applicable): NA
Additional Context
This is not an issue, if the current experiment is ended and new run_evals is executed in a different experiment
Error Logs
NA