Open hmaarrfk opened 3 days ago
The comment https://github.com/pydata/xarray/pull/9002/files#r1591214299 had the right intuition. The fastpath was the source of problems.
Empirically, i tested to see if it had any effects on the internal benchmarks that brought me to introducing it and it seems it does not (we avoid pandas quite heavily due to slowdowns it introduces, our data access is quite regular).
ps. i have visitors this week, so feel free to push to adjust anything you feel the need to change.
Not saying anything should change here, just giving the results of the benchmark
asv continuous main calendar_noleap --bench indexing
| Change | Before [42ed6d30] <main> | After [822c0b8e] <calendar_noleap> | Ratio | Benchmark (Parameter) |
|----------|----------------------------|--------------------------------------|---------|------------------------------------------------------------|
| + | 337±1μs | 618±1μs | 1.83 | indexing.AssignmentOptimized.time_assign_identical_indexes |
| + | 67.4±0.6μs | 77.7±0.5μs | 1.15 | indexing.HugeAxisSmallSliceIndexing.time_indexing |
| + | 139±0.7μs | 159±0.5μs | 1.14 | indexing.Assignment.time_assignment_basic('1slice') |
@pydata/xarray who is the best person to review this?
I believe the solution in https://github.com/pydata/xarray/issues/9138#issuecomment-2198145554 is a better path
See discussion in #9138 Fixes #9138
This commit and pull request mostly serves as a staging group for a potential fix.
Test with:
whats-new.rst
api.rst