Skip to content

perf: parallel downsample #4004

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
flying-sheep wants to merge 43 commits into
base: main
Choose a base branch
from

Merge branch 'pa/rng' into parallel-downsample

d748f0b
Select commit
Loading
Failed to load commit list.
Draft

perf: parallel downsample #4004

Merge branch 'pa/rng' into parallel-downsample
d748f0b
Select commit
Loading
Failed to load commit list.
scverse-benchmark / benchmark succeeded Mar 20, 2026 in 37m 35s

Benchmark

Benchmark run successful

Details

All benchmarks:

Change Before [9de11b1] After [d748f0b] Ratio Benchmark (Parameter)
311M 310M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('bmmc', 'counts')
311M 311M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('bmmc', 'counts-off-axis')
4.11G 4.11G 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('lung93k', 'counts')
4.11G 4.11G 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('lung93k', 'counts-off-axis')
378M 378M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('pbmc3k', 'counts')
377M 377M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('pbmc3k', 'counts-off-axis')
288M 289M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('pbmc68k_reduced', 'counts')
289M 289M 1.00 preprocessing_counts.FastSuite.peakmem_calculate_qc_metrics('pbmc68k_reduced', 'counts-off-axis')
314M 313M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('bmmc', 'counts')
313M 313M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('bmmc', 'counts-off-axis')
4.45G 4.45G 1.00 preprocessing_counts.FastSuite.peakmem_log1p('lung93k', 'counts')
4.45G 4.45G 1.00 preprocessing_counts.FastSuite.peakmem_log1p('lung93k', 'counts-off-axis')
384M 385M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('pbmc3k', 'counts')
384M 385M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('pbmc3k', 'counts-off-axis')
288M 289M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('pbmc68k_reduced', 'counts')
288M 288M 1.00 preprocessing_counts.FastSuite.peakmem_log1p('pbmc68k_reduced', 'counts-off-axis')
402M 402M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('bmmc', 'counts')
401M 401M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('bmmc', 'counts-off-axis')
4.96G 4.96G 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('lung93k', 'counts')
4.97G 4.96G 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('lung93k', 'counts-off-axis')
474M 474M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('pbmc3k', 'counts')
473M 474M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('pbmc3k', 'counts-off-axis')
288M 287M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('pbmc68k_reduced', 'counts')
288M 288M 1.00 preprocessing_counts.FastSuite.peakmem_normalize_total('pbmc68k_reduced', 'counts-off-axis')
12.6±0.1ms 12.6±0.3ms 1.00 preprocessing_counts.FastSuite.time_calculate_qc_metrics('bmmc', 'counts')
12.5±0.1ms 12.5±0.3ms 0.99 preprocessing_counts.FastSuite.time_calculate_qc_metrics('bmmc', 'counts-off-axis')
2.06±0.01s 2.06±0.01s 1.00 preprocessing_counts.FastSuite.time_calculate_qc_metrics('lung93k', 'counts')
1.64±0s 1.63±0.01s 1.00 preprocessing_counts.FastSuite.time_calculate_qc_metrics('lung93k', 'counts-off-axis')
38.1±0.7ms 37.9±0.3ms 0.99 preprocessing_counts.FastSuite.time_calculate_qc_metrics('pbmc3k', 'counts')
28.8±3ms 27.8±0.4ms 0.96 preprocessing_counts.FastSuite.time_calculate_qc_metrics('pbmc3k', 'counts-off-axis')
4.70±0.05ms 4.78±0.03ms 1.02 preprocessing_counts.FastSuite.time_calculate_qc_metrics('pbmc68k_reduced', 'counts')
4.72±0.04ms 4.71±0.03ms 1.00 preprocessing_counts.FastSuite.time_calculate_qc_metrics('pbmc68k_reduced', 'counts-off-axis')
1.53±0.02ms 1.54±0.06ms 1.01 preprocessing_counts.FastSuite.time_log1p('bmmc', 'counts')
1.58±0.04ms 1.59±0.01ms 1.01 preprocessing_counts.FastSuite.time_log1p('bmmc', 'counts-off-axis')
638±2ms 639±4ms 1.00 preprocessing_counts.FastSuite.time_log1p('lung93k', 'counts')
637±3ms 630±2ms 0.99 preprocessing_counts.FastSuite.time_log1p('lung93k', 'counts-off-axis')
7.15±0.03ms 7.11±0.2ms 0.99 preprocessing_counts.FastSuite.time_log1p('pbmc3k', 'counts')
7.30±0.06ms 7.59±0.4ms 1.04 preprocessing_counts.FastSuite.time_log1p('pbmc3k', 'counts-off-axis')
387±4μs 386±0.6μs 1.00 preprocessing_counts.FastSuite.time_log1p('pbmc68k_reduced', 'counts')
420±30μs 398±7μs 0.95 preprocessing_counts.FastSuite.time_log1p('pbmc68k_reduced', 'counts-off-axis')
2.69±0.2ms 2.64±0.04ms 0.98 preprocessing_counts.FastSuite.time_normalize_total('bmmc', 'counts')
6.65±0.4ms 6.76±0.3ms 1.02 preprocessing_counts.FastSuite.time_normalize_total('bmmc', 'counts-off-axis')
548±7ms 551±10ms 1.01 preprocessing_counts.FastSuite.time_normalize_total('lung93k', 'counts')
2.71±0.03s 2.75±0.04s 1.01 preprocessing_counts.FastSuite.time_normalize_total('lung93k', 'counts-off-axis')
8.72±0.5ms 8.69±0.2ms 1.00 preprocessing_counts.FastSuite.time_normalize_total('pbmc3k', 'counts')
33.9±1ms 33.7±1ms 0.99 preprocessing_counts.FastSuite.time_normalize_total('pbmc3k', 'counts-off-axis')
546±0.9μs 549±4μs 1.01 preprocessing_counts.FastSuite.time_normalize_total('pbmc68k_reduced', 'counts')
547±0.5μs 548±2μs 1.00 preprocessing_counts.FastSuite.time_normalize_total('pbmc68k_reduced', 'counts-off-axis')
399M 420M 1.05 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_per_cell('pbmc3k', 'random_state')
399M 420M 1.05 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_per_cell('pbmc3k', 'rng')
336M 337M 1.00 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_per_cell('pbmc68k_reduced', 'random_state')
335M 349M 1.04 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_per_cell('pbmc68k_reduced', 'rng')
468M 474M 1.01 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_total('pbmc3k', 'random_state')
+ 431M 477M 1.11 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_total('pbmc3k', 'rng')
340M 342M 1.01 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_total('pbmc68k_reduced', 'random_state')
338M 346M 1.02 preprocessing_counts.PreprocessingCountsRngSuite.peakmem_downsample_total('pbmc68k_reduced', 'rng')
- 205±0.5ms 28.0±6ms 0.14 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_per_cell('pbmc3k', 'random_state')
- 71.8±0.7ms 20.8±6ms 0.29 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_per_cell('pbmc3k', 'rng')
- 25.1±0.2ms 1.48±0.2ms 0.06 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_per_cell('pbmc68k_reduced', 'random_state')
16.1±0.08ms 3.67±0.06ms ~0.23 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_per_cell('pbmc68k_reduced', 'rng')
+ 192±8ms 306±4ms 1.59 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_total('pbmc3k', 'random_state')
+ 63.3±0.3ms 182±20ms 2.88 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_total('pbmc3k', 'rng')
+ 13.1±0.4ms 16.7±2ms 1.27 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_total('pbmc68k_reduced', 'random_state')
+ 7.14±0.3ms 11.1±0.6ms 1.55 preprocessing_counts.PreprocessingCountsRngSuite.time_downsample_total('pbmc68k_reduced', 'rng')
431M 430M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_cells('pbmc3k', 'counts')
431M 431M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_cells('pbmc3k', 'counts-off-axis')
299M 300M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_cells('pbmc68k_reduced', 'counts')
299M 299M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_cells('pbmc68k_reduced', 'counts-off-axis')
431M 431M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_genes('pbmc3k', 'counts')
430M 431M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_genes('pbmc3k', 'counts-off-axis')
299M 299M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_genes('pbmc68k_reduced', 'counts')
299M 299M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_filter_genes('pbmc68k_reduced', 'counts-off-axis')
1.1G 1.11G 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_scrublet('pbmc3k', 'counts')
1.1G 1.1G 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_scrublet('pbmc3k', 'counts-off-axis')
498M 524M 1.05 preprocessing_counts.PreprocessingCountsSuite.peakmem_scrublet('pbmc68k_reduced', 'counts')
522M 523M 1.00 preprocessing_counts.PreprocessingCountsSuite.peakmem_scrublet('pbmc68k_reduced', 'counts-off-axis')
57.2±8ms 56.2±0.9ms 0.98 preprocessing_counts.PreprocessingCountsSuite.time_filter_cells('pbmc3k', 'counts')
58.3±1ms 58.8±3ms 1.01 preprocessing_counts.PreprocessingCountsSuite.time_filter_cells('pbmc3k', 'counts-off-axis')
10.2±0.8ms 10.4±0.9ms 1.02 preprocessing_counts.PreprocessingCountsSuite.time_filter_cells('pbmc68k_reduced', 'counts')
12.4±0.9ms 9.71±0.9ms ~0.78 preprocessing_counts.PreprocessingCountsSuite.time_filter_cells('pbmc68k_reduced', 'counts-off-axis')
52.6±1ms 52.2±0.5ms 0.99 preprocessing_counts.PreprocessingCountsSuite.time_filter_genes('pbmc3k', 'counts')
52.0±3ms 50.8±3ms 0.98 preprocessing_counts.PreprocessingCountsSuite.time_filter_genes('pbmc3k', 'counts-off-axis')
10.5±0.9ms 10.8±0.9ms 1.03 preprocessing_counts.PreprocessingCountsSuite.time_filter_genes('pbmc68k_reduced', 'counts')
10.5±0.9ms 10.8±0.9ms 1.03 preprocessing_counts.PreprocessingCountsSuite.time_filter_genes('pbmc68k_reduced', 'counts-off-axis')
2.27±0.2s 2.60±0.1s ~1.15 preprocessing_counts.PreprocessingCountsSuite.time_scrublet('pbmc3k', 'counts')
2.32±0.2s 2.59±0.05s ~1.12 preprocessing_counts.PreprocessingCountsSuite.time_scrublet('pbmc3k', 'counts-off-axis')
559±8ms 546±9ms 0.98 preprocessing_counts.PreprocessingCountsSuite.time_scrublet('pbmc68k_reduced', 'counts')
554±9ms 549±8ms 0.99 preprocessing_counts.PreprocessingCountsSuite.time_scrublet('pbmc68k_reduced', 'counts-off-axis')
432M 432M 1.00 preprocessing_log.PreprocessingSuite.peakmem_highly_variable_genes('pbmc3k', 'off-axis')
480M 480M 1.00 preprocessing_log.PreprocessingSuite.peakmem_highly_variable_genes('pbmc3k', None)
296M 296M 1.00 preprocessing_log.PreprocessingSuite.peakmem_highly_variable_genes('pbmc68k_reduced', 'off-axis')
297M 297M 1.00 preprocessing_log.PreprocessingSuite.peakmem_highly_variable_genes('pbmc68k_reduced', None)
569M 568M 1.00 preprocessing_log.PreprocessingSuite.peakmem_pca('pbmc3k', 'off-axis')
588M 588M 1.00 preprocessing_log.PreprocessingSuite.peakmem_pca('pbmc3k', None)
492M 496M 1.01 preprocessing_log.PreprocessingSuite.peakmem_pca('pbmc68k_reduced', 'off-axis')
498M 498M 1.00 preprocessing_log.PreprocessingSuite.peakmem_pca('pbmc68k_reduced', None)
n/a n/a n/a preprocessing_log.PreprocessingSuite.peakmem_regress_out('pbmc3k', 'off-axis')
n/a n/a n/a preprocessing_log.PreprocessingSuite.peakmem_regress_out('pbmc3k', None)
351M 349M 0.99 preprocessing_log.PreprocessingSuite.peakmem_regress_out('pbmc68k_reduced', 'off-axis')
350M 351M 1.00 preprocessing_log.PreprocessingSuite.peakmem_regress_out('pbmc68k_reduced', None)
1.29G 1.29G 1.00 preprocessing_log.PreprocessingSuite.peakmem_scale('pbmc3k', 'off-axis')
1.49G 1.49G 1.00 preprocessing_log.PreprocessingSuite.peakmem_scale('pbmc3k', None)
336M 341M 1.01 preprocessing_log.PreprocessingSuite.peakmem_scale('pbmc68k_reduced', 'off-axis')
338M 343M 1.01 preprocessing_log.PreprocessingSuite.peakmem_scale('pbmc68k_reduced', None)
39.2±2ms 39.5±2ms 1.01 preprocessing_log.PreprocessingSuite.time_highly_variable_genes('pbmc3k', 'off-axis')
55.8±0.5ms 55.3±0.6ms 0.99 preprocessing_log.PreprocessingSuite.time_highly_variable_genes('pbmc3k', None)
16.8±0.2ms 16.6±0.03ms 0.99 preprocessing_log.PreprocessingSuite.time_highly_variable_genes('pbmc68k_reduced', 'off-axis')
16.8±0.2ms 16.6±0.2ms 0.99 preprocessing_log.PreprocessingSuite.time_highly_variable_genes('pbmc68k_reduced', None)
1.96±0.01s 1.94±0.02s 0.99 preprocessing_log.PreprocessingSuite.time_pca('pbmc3k', 'off-axis')
2.15±0.01s 2.16±0.01s 1.00 preprocessing_log.PreprocessingSuite.time_pca('pbmc3k', None)
170±10ms 166±20ms 0.98 preprocessing_log.PreprocessingSuite.time_pca('pbmc68k_reduced', 'off-axis')
104±50ms 166±3ms ~1.60 preprocessing_log.PreprocessingSuite.time_pca('pbmc68k_reduced', None)
n/a n/a n/a preprocessing_log.PreprocessingSuite.time_regress_out('pbmc3k', 'off-axis')
n/a n/a n/a preprocessing_log.PreprocessingSuite.time_regress_out('pbmc3k', None)
16.9±0.8ms 17.0±0.7ms 1.01 preprocessing_log.PreprocessingSuite.time_regress_out('pbmc68k_reduced', 'off-axis')
16.7±0.4ms 16.9±0.6ms 1.02 preprocessing_log.PreprocessingSuite.time_regress_out('pbmc68k_reduced', None)
499±1ms 497±2ms 1.00 preprocessing_log.PreprocessingSuite.time_scale('pbmc3k', 'off-axis')
554±1ms 551±1ms 0.99 preprocessing_log.PreprocessingSuite.time_scale('pbmc3k', None)
4.72±0.2ms 4.76±0.1ms 1.01 preprocessing_log.PreprocessingSuite.time_scale('pbmc68k_reduced', 'off-axis')
4.61±0.1ms 4.74±0.1ms 1.03 preprocessing_log.PreprocessingSuite.time_scale('pbmc68k_reduced', None)
287M 287M 1.00 tools.ToolsSuite.peakmem_diffmap
293M 294M 1.00 tools.ToolsSuite.peakmem_leiden
362M 362M 1.00 tools.ToolsSuite.peakmem_rank_genes_groups
451M 449M 1.00 tools.ToolsSuite.peakmem_umap
16.7±0.4ms 16.5±0.3ms 0.99 tools.ToolsSuite.time_diffmap
20.1±0.2ms 19.5±0.07ms 0.97 tools.ToolsSuite.time_leiden
64.4±5ms 60.3±5ms 0.94 tools.ToolsSuite.time_rank_genes_groups
1.02±0s 1.02±0s 1.00 tools.ToolsSuite.time_umap
View more details on scverse-benchmark