v0.9.3

What's Changed

• Fix direct reductions of Xarray objects by @dcherian in #339

Full Changelog: v0.9.2...v0.9.3

v0.9.2

Speed up the new quantile reduction

What's Changed

• quantiles use complex array partition, not lexsort by @dcherian in #335
• Test with py3.12 by @dcherian in #336

Full Changelog: v0.9.1...v0.9.2

v0.9.1

Vectorized grouped quantiles!

What's Changed

• Vectorized grouped (nan)quantile by @dcherian in #329
• Pass ddof through for numbagg by @dcherian in #302
• Silence warnings by @dcherian in #333
• Support multiple quantiles with xarray by @dcherian in #332

Full Changelog: v0.9.0...v0.9.1

v0.9.0

What's Changed

method=None by default, and we have heuristics for setting it appropriately!

• Heuristics for auto-guessing "method" by @dcherian in #314
• Optimize bitmask finding some more. by @dcherian in #326
• Add docs on method=None heuristics by @dcherian in #328

Full Changelog: v0.8.9...v0.9.0

v0.8.9

Fix dependencies

v0.8.8

What's Changed

• Support scipy < 1.11

Full Changelog: v0.8.7...v0.8.8

v0.8.7

Bugfix release

What's Changed

• Cleanup by @dcherian in #315
• Bugfix for cohorts where not all expected_groups are present by @dcherian in #316

Full Changelog: v0.8.6...v0.8.7

v0.8.6

What's Changed

More cohorts optimization + a bugfix.

• Optimize cohorts for blockwise distribution of groups by @dcherian in #299
• Use dot product for containment by @dcherian in #306
• Fix missing resampling groups. by @dcherian in #312

Full Changelog: v0.8.5...v0.8.6

v0.8.5

Add scipy as required dependency

See 0.8.4 release notes

v0.8.4

What's Changed

Another round of absolutely massive performance improvements for method="cohorts". This should bigly improve many Xarray groupby workloads (which use cohorts by default). Resampling in particular should be much better

Benchmarks improvements undersell the changes, since the core loop is approximately quadratic. Graph construction time for the example in https://xarray.dev/blog/flox with "cohorts" specified now drops from 30s to 3s 😱

| Before [15324a7a] <v0.8.3>   | After [666d45e9] <main>   |   Ratio | Benchmark (Parameter)                                   |
|------------------------------|---------------------------|---------|---------------------------------------------------------|

#### Larger cohorts, lower tasks
| 5180                         | 3600                      |    0.69 | cohorts.NWMMidwest.track_num_tasks                      |
| 4891                         | 3385                      |    0.69 | cohorts.NWMMidwest.track_num_tasks_optimized            |
| 505                          | 345                       |    0.68 | cohorts.NWMMidwest.track_num_layers                     |

#### Much faster algorithm for detecting cohorts, that should scale better.
| 3.19±0.07ms                  | 2.42±0.05ms               |    0.76 | cohorts.ERA5Google.time_find_group_cohorts              |
| 1.04±0.01ms                  | 782±70μs                  |    0.75 | cohorts.PerfectMonthly.time_find_group_cohorts          |
| 1.06±0.01ms                  | 781±70μs                  |    0.73 | cohorts.PerfectMonthlyRechunked.time_find_group_cohorts |
| 29.7±2ms                     | 12.6±0.9ms                |    0.43 | cohorts.ERA5DayOfYear.time_find_group_cohorts           |
| 7.76±1ms                     | 2.90±0.2ms                |    0.37 | cohorts.ERA5MonthHour.time_find_group_cohorts           |
| 8.17±0.8ms                   | 2.75±0.2ms                |    0.34 | cohorts.ERA5MonthHourRechunked.time_find_group_cohorts  |
| 242±5ms                      | 47.3±2ms                  |    0.2  | cohorts.NWMMidwest.time_find_group_cohorts              |
| 28.8±3ms                     | 4.11±0.3ms                |    0.14 | cohorts.ERA5DayOfYearRechunked.time_find_group_cohorts  |

##### Total time is not too different, we have some overhead in constructing the graphs
| 162±5ms                      | 144±9ms                   |    0.89 | cohorts.ERA5DayOfYearRechunked.time_graph_construct     |
| 20.7±0.2ms                   | 18.3±0.4ms                |    0.89 | cohorts.ERA5Google.time_graph_construct                 |
| 3.21±0.2ms                   | 2.40±0.04ms               |    0.75 | cohorts.PerfectMonthly.time_graph_construct             |
| 181±10ms                     | 129±10ms                  |    0.71 | cohorts.NWMMidwest.time_graph_construct                 |

Changes

• More cohorts speedups by @dcherian in #290
• typing fixes. by @dcherian in #292
• Use set containment instead of perfect subsets by @dcherian in #291

Full Changelog: v0.8.3...v0.8.4