Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* adding experimental jupyter notebook in experimental/ * updated all calls and nu test to include loss method arguments * removed invalid includes from collection and directly added to chassis.py; * flake8 configured to ignore E501(some lines longer than 80) and E203(conflicts with black) * moved chassis loop to MuyGPyS/optimize/experimental and updated __init__.py * add notebook for distance function testing * added updating neighborhood with learned length scales each loop iteration * add parameter and code that allows reuse of optimizer instance over loop iterations * added test for anisotropic distance function * updated chassis workflow when reusing optimizer and update notebook with experiment and 4 treatments * returning the nn lookup for inference * added workflow tests determining where learned hps applied in opt loop and troubleshooting inference results * added inference outputs test * added training and inference results to tables in notebook * refactored sample_batch call to to make workflow understandable, added time and probe count measurements, and added tables for training and inference results and a side by side plot of residuals * added Abseil unit test to github CI * Added another treatment single step + that computes an equal number of explore and exploit points
- Loading branch information