automl / automl/ConfigurableOptimizer
Are we tracking the correct layer alignment scores?
Open
High priority
old
- Dominant language
- Python
- Stars
- 12
- Forks
- 2
- PR merge metrics
- No merged PRs in 30d
Description
In the [Lambda-DARTS repo](https://github.com/dr-faustus/Lambda-DARTS/blob/master/NAS-Bench201/exps/algos/DARTS-V1.py), they track a `train_corr` and a `valid_corr`.
What we're tracking looks equivalent to `train_corr`. Is this really the one we want to track?
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.