This library provides a framework for assessing wind plant performance using operational assessment (OA) methodologies that consume time series data from wind plants. The goal of the project is to provide an open source implementation of common data structures, analysis methods, and utility functions relevant to wind plant OA.
In https://github.com/NREL/OpenOA/pull/226 , we tried to understand why certain regression tests failed when updating Pandas and Scikit-Learn. We were unable to fully identify the root cause and made a decision at the time to simply updated the regression tests for now. Before the release is made, it is imperative to better describe and understand the root cause of these regressions, and update the AEP GBM and Wake Losses modules so it doesn't happen again in the future.
Understand and describe the root cause of the regressions.
Update the affected analysis modules.
Add unit tests to ensure regressions like this are easier to debug in the future.
At this stage, I believe we've just accepted the differences with no pushback on results discrepancies, so I'm closing this, though the stability of these tests has long been an unresolved issue.
RHammond2
commented
5 months ago
In https://github.com/NREL/OpenOA/pull/226 , we tried to understand why certain regression tests failed when updating Pandas and Scikit-Learn. We were unable to fully identify the root cause and made a decision at the time to simply updated the regression tests for now. Before the release is made, it is imperative to better describe and understand the root cause of these regressions, and update the AEP GBM and Wake Losses modules so it doesn't happen again in the future.