We are observing testthat test failures when building harmony (v1.2.3) in Conda Forge. Specifically, we see results like:
── Failure ('test_integration.R:38:5'): increasing theta decreases chi2 between Cluster and Batch assign ──
Error: Test failures
sum(((obj0$O - obj0$E)^2)/obj0$E) is not strictly more than sum(((obj1$O - obj1$E)^2)/obj1$E). Difference: -53.8
So far, this has only be observed during win-64 platform runs. Different runs both failed and passed in R 4.3 and 4.4 builds. That is, it appears sporadic.
I haven't looked any further into this - just bubbling up what we're seeing downstream.
We are observing testthat test failures when building
harmony(v1.2.3) in Conda Forge. Specifically, we see results like:So far, this has only be observed during
win-64platform runs. Different runs both failed and passed in R 4.3 and 4.4 builds. That is, it appears sporadic.I haven't looked any further into this - just bubbling up what we're seeing downstream.