nightly_run - CyrilB1531/lodestar GitHub Wiki

Nightly benchmark run

Generated. Do not edit. Produced by .github/workflows/bench-nightly.yml; every edit is overwritten by the next run. The curated figures, measured on a named machine, are in performance. The last known reading for a method quiet tonight is in benchmark_latest.

Read the ratios, not the means. These run on a GitHub hosted runner: a shared VM whose hardware differs from night to night and whose neighbours are unknown. An absolute figure here is not comparable to the performance page, and not reliably comparable to yesterday's. A ratio against a baseline measured in the same run, on the same VM, in the same minute, is.

This run

Commit: fac280c117f7e86c65fd70bf1cf12e07b683e38a
Previous run: fac280c117f7e86c65fd70bf1cf12e07b683e38a
Runner: Linux / X64 (GitHub hosted)

Classes re-run

Selected by tools/select_benchmarks.py from the sources that changed since the previous run:

BatchEmbeddingBenchmarks
BkTreeBenchmarks
BlockedTableBenchmarks
BpeBenchmarks
BpeScalingBenchmarks
BpeWordCacheBenchmarks
BucketRouteDiagnostics
DecompositionBenchmarks
KMeansFitIncumbentBenchmarks
KMeansLloydIncumbentBenchmarks
DbscanIncumbentBenchmarks
DbscanDimensionBenchmarks
AgglomerativeIncumbentBenchmarks
PartialFitBenchmarks
PrincipalComponentVarianceBenchmarks
FuzzBenchmarks
FuzzIncumbentBenchmarks
FuzzCodePointBenchmarks
MetaNumericsPcaBenchmarks
MetaNumericsStatsBenchmarks
KsAutoBenchmarks
MetricsIncumbentBenchmarks
TokenizerIncumbentBenchmarks
VectorizerIncumbentBenchmarks
IndelBenchmarks
IndelCodePointBenchmarks
LcsGateBenchmarks
LevenshteinBenchmarks
LevenshteinIncumbentBenchmarks
LevenshteinCodePointBenchmarks
MetricsBenchmarks
RegressionMetricsBenchmarks
MyersGateBenchmarks
PartialRatioLongNeedleBenchmarks
PersistenceBenchmarks
StopWordBenchmarks
VectorMathBenchmarks
VectorizerBenchmarks
SimilaritySketchBenchmarks
Bm25Benchmarks
CoxBenchmarks
SurvivalBenchmarks
TiledCosineTopKBenchmarks
TiledSparseDenseProductBenchmarks
BitParallelEditDistanceBenchmarks
ChainedProductBenchmarks
TiledMinHashSignaturesBenchmarks
StatsBenchmarks
RankTestBenchmarks
DistributionTailBenchmarks
SplitterIncumbentBenchmarks
StationarityBenchmarks
EncoderIncumbentBenchmarks
ScalerIncumbentBenchmarks
SerialCorrelationBenchmarks
QuantileBenchmarks
OlsBenchmarks
GlsBenchmarks
WeightedLeastSquaresBenchmarks
RobustCovarianceBenchmarks
HacClusterBenchmarks
GlmBenchmarks
GlmPoissonBenchmarks
GlmOffsetBenchmarks
MultinomialLogitBenchmarks
VectorAutoregressionBenchmarks
TextRankBenchmarks
SilhouetteBenchmarks
ClusteringAgreementBenchmarks
EmbeddingSearchBenchmarks
MannWhitneyExactBenchmarks
RobustScalerSparseBenchmarks
TopKAccuracyBenchmarks
DamerauLevenshteinBenchmarks
QgramBenchmarks
KsExactTableBenchmarks
OsaBenchmarks
RatcliffObershelpBenchmarks
DoubleMetaphoneBenchmarks
ProcessExtractBenchmarks
FilteredVectorSearchBenchmarks
SentencePieceBpeLineageBenchmarks
BertNormalizerBenchmarks
AddedTokenScanBenchmarks
PrecompiledNormalizerBenchmarks
EmbeddingSearchHelpersBenchmarks
RankingMetricsBenchmarks
MultilabelConfusionMatrixBenchmarks
ClassifierCurveBenchmarks
PartitionValidityBenchmarks
ClassificationReportBenchmarks
MultiClassRocAucBenchmarks
KolmogorovDurbinBenchmarks
DecompositionWidthBenchmarks
HouseholderQrBenchmarks
GlmNegativeBinomialBenchmarks
LeastSquaresRoutingBenchmarks

Lodestar.Stats.Benchmarks.DistributionTailBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
ChiSquaredSfOneDf	18.454 ns	3.6493 ns	0.2000 ns	1.00	0.01	-	NA
ChiSquaredSfFourDf	6.992 ns	0.2986 ns	0.0164 ns	0.38	0.00	-	NA
ChiSquaredSfThreeDfFarTail	24.646 ns	8.5655 ns	0.4695 ns	1.34	0.03	-	NA
ChiSquaredSfHundredDf	53.795 ns	15.3009 ns	0.8387 ns	2.92	0.05	-	NA
ChiSquaredSfFractionalDf	62.592 ns	30.8356 ns	1.6902 ns	3.39	0.09	-	NA
NormalQuantile	94.382 ns	3.8566 ns	0.2114 ns	5.11	0.05	-	NA

Lodestar.Stats.Benchmarks.KsAutoBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleSize	Mean	Error	StdDev	Ratio	Gen0	Allocated	Alloc Ratio
Before_Asymptotic	1000	55.39 μs	5.156 μs	0.283 μs	1.00	0.1831	15.73 KB	1.00
After_Auto	1000	52.08 μs	11.331 μs	0.621 μs	0.94	0.1831	15.67 KB	1.00

Before_Asymptotic	10000	1,517.53 μs	39.269 μs	2.152 μs	1.00	-	156.3 KB	1.00
After_Auto	10000	1,264.22 μs	244.318 μs	13.392 μs	0.83	-	156.3 KB	1.00

Lodestar.Stats.Benchmarks.MetaNumericsStatsBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleSize	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Lodestar_StudentT	100	539.99 ns	4.945 ns	0.271 ns	1.00	0.00	-	-	NA
MetaNumerics_StudentT	100	1,702.48 ns	36.437 ns	1.997 ns	3.15	0.00	-	104 B	NA
Lodestar_MannWhitney	100	2,193.56 ns	68.850 ns	3.774 ns	4.06	0.01	-	-	NA
MetaNumerics_MannWhitney	100	3,251.33 ns	871.926 ns	47.793 ns	6.02	0.08	0.0114	1176 B	NA
Lodestar_KruskalWallis	100	5,437.69 ns	558.770 ns	30.628 ns	10.07	0.05	-	32 B	NA
MetaNumerics_KruskalWallis	100	7,032.80 ns	2,617.391 ns	143.468 ns	13.02	0.23	0.0153	1896 B	NA
Lodestar_KolmogorovSmirnov	100	2,163.05 ns	159.473 ns	8.741 ns	4.01	0.01	0.0191	1648 B	NA
MetaNumerics_KolmogorovSmirnov	100	3,243.11 ns	1,222.371 ns	67.002 ns	6.01	0.11	0.0114	1168 B	NA
Lodestar_OneWayAnova	100	571.19 ns	15.745 ns	0.863 ns	1.06	0.00	-	32 B	NA
MetaNumerics_OneWayAnova	100	2,345.64 ns	120.530 ns	6.607 ns	4.34	0.01	0.0076	744 B	NA
Lodestar_Wilcoxon	100	988.28 ns	15.936 ns	0.874 ns	1.83	0.00	0.0095	824 B	NA
MetaNumerics_Wilcoxon	100	1,024.87 ns	555.239 ns	30.434 ns	1.90	0.05	0.0114	976 B	NA
Lodestar_FisherExact	100	264.57 ns	161.935 ns	8.876 ns	0.49	0.01	-	-	NA
MetaNumerics_FisherExact	100	1,115.75 ns	244.696 ns	13.413 ns	2.07	0.02	0.0095	944 B	NA
Lodestar_ChiSquare	100	88.82 ns	13.823 ns	0.758 ns	0.16	0.00	0.0019	168 B	NA
MetaNumerics_ChiSquare	100	556.44 ns	102.908 ns	5.641 ns	1.03	0.01	0.0114	968 B	NA

Lodestar_StudentT	10000	23,074.69 ns	9,763.196 ns	535.154 ns	1.000	0.03	-	-	NA
MetaNumerics_StudentT	10000	140,915.39 ns	13,737.395 ns	752.993 ns	6.109	0.12	-	104 B	NA
Lodestar_MannWhitney	10000	441,043.40 ns	3,017.042 ns	165.374 ns	19.120	0.38	-	-	NA
MetaNumerics_MannWhitney	10000	1,772,657.53 ns	104,293.541 ns	5,716.681 ns	76.850	1.54	-	80379 B	NA
Lodestar_KruskalWallis	10000	1,001,039.09 ns	118,234.095 ns	6,480.810 ns	43.398	0.89	-	35 B	NA
MetaNumerics_KruskalWallis	10000	3,112,775.39 ns	1,366,615.458 ns	74,908.805 ns	134.948	3.88	-	120702 B	NA
Lodestar_KolmogorovSmirnov	10000	1,240,560.07 ns	74,694.849 ns	4,094.277 ns	53.782	1.08	-	160051 B	NA
MetaNumerics_KolmogorovSmirnov	10000	1,787,746.94 ns	91,732.793 ns	5,028.184 ns	77.504	1.55	-	80371 B	NA
Lodestar_OneWayAnova	10000	51,232.48 ns	15,921.842 ns	872.730 ns	2.221	0.05	-	32 B	NA
MetaNumerics_OneWayAnova	10000	212,166.12 ns	30,835.509 ns	1,690.198 ns	9.198	0.19	-	744 B	NA
Lodestar_Wilcoxon	10000	215,935.62 ns	31,532.313 ns	1,728.392 ns	9.361	0.20	0.7324	80056 B	NA
MetaNumerics_Wilcoxon	10000	369,656.25 ns	19,765.774 ns	1,083.429 ns	16.026	0.32	0.4883	80177 B	NA
Lodestar_FisherExact	10000	256.10 ns	6.269 ns	0.344 ns	0.011	0.00	-	-	NA
MetaNumerics_FisherExact	10000	1,099.53 ns	1.897 ns	0.104 ns	0.048	0.00	0.0095	944 B	NA
Lodestar_ChiSquare	10000	88.71 ns	9.226 ns	0.506 ns	0.004	0.00	0.0019	168 B	NA
MetaNumerics_ChiSquare	10000	552.20 ns	163.898 ns	8.984 ns	0.024	0.00	0.0114	968 B	NA

Lodestar.Stats.Benchmarks.RankTestBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleSize	Ties	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
KruskalWallisTest	10000	False	1,053.1 μs	50.38 μs	2.76 μs	-	-	-	82 B
WilcoxonPaired	10000	False	271.3 μs	25.37 μs	1.39 μs	0.4883	-	-	80057 B
KruskalWallisManyGroups	10000	False	585.0 μs	20.97 μs	1.15 μs	2.9297	-	-	279681 B
WilcoxonPooledDifferences	10000	False	885.4 μs	194.55 μs	10.66 μs	-	-	-	-
MannWhitneyTest	10000	False	472.1 μs	25.10 μs	1.38 μs	-	-	-	-
KruskalWallisTest	10000	True	647.4 μs	47.38 μs	2.60 μs	-	-	-	81 B
WilcoxonPaired	10000	True	246.2 μs	7.32 μs	0.40 μs	0.4883	-	-	80057 B
KruskalWallisManyGroups	10000	True	406.4 μs	6.41 μs	0.35 μs	2.9297	0.4883	-	279680 B
WilcoxonPooledDifferences	10000	True	796.7 μs	358.34 μs	19.64 μs	-	-	-	-
MannWhitneyTest	10000	True	440.9 μs	118.62 μs	6.50 μs	-	-	-	-
KruskalWallisTest	100000	False	11,085.7 μs	4,716.03 μs	258.50 μs	-	-	-	-
WilcoxonPaired	100000	False	3,249.2 μs	141.91 μs	7.78 μs	113.2813	113.2813	113.2813	800172 B
KruskalWallisManyGroups	100000	False	8,077.0 μs	3,123.19 μs	171.19 μs	328.1250	328.1250	328.1250	2800349 B
WilcoxonPooledDifferences	100000	False	28,339.3 μs	1,659.90 μs	90.98 μs	968.7500	968.7500	968.7500	13200772 B
MannWhitneyTest	100000	False	4,923.1 μs	116.42 μs	6.38 μs	-	-	-	-
KruskalWallisTest	100000	True	6,471.1 μs	141.59 μs	7.76 μs	-	-	-	88 B
WilcoxonPaired	100000	True	2,847.9 μs	105.20 μs	5.77 μs	113.2813	113.2813	113.2813	800172 B
KruskalWallisManyGroups	100000	True	4,722.4 μs	982.16 μs	53.84 μs	390.6250	390.6250	390.6250	2800386 B
WilcoxonPooledDifferences	100000	True	18,215.2 μs	1,637.67 μs	89.77 μs	968.7500	968.7500	968.7500	13076300 B
MannWhitneyTest	100000	True	4,349.9 μs	594.41 μs	32.58 μs	-	-	-	-

Lodestar.Stats.Benchmarks.StationarityBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleSize	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
LodestarAugmentedDickeyFuller	200	6.733 μs	6.7601 μs	0.3705 μs	1.00	0.07	0.2899	-	-	23.99 KB	1.00
LodestarAdfAutolag	200	24.819 μs	5.5951 μs	0.3067 μs	3.69	0.18	0.7935	0.0610	-	66.56 KB	2.77
CortexAugmentedDickeyFuller	200	22.372 μs	0.4249 μs	0.0233 μs	3.33	0.15	0.1831	-	-	16.18 KB	0.67
LodestarKpss	200	2.791 μs	0.9966 μs	0.0546 μs	0.42	0.02	0.0191	-	-	1.69 KB	0.07
CortexKpss	200	2.867 μs	1.0379 μs	0.0569 μs	0.43	0.02	0.0420	-	-	3.43 KB	0.14
LodestarDecompose	200	1.312 μs	0.3312 μs	0.0182 μs	0.20	0.01	0.0782	-	-	6.5 KB	0.27
CortexDecompose	200	2.679 μs	0.3479 μs	0.0191 μs	0.40	0.02	0.0992	-	-	8.16 KB	0.34

LodestarAugmentedDickeyFuller	2000	86.184 μs	6.4010 μs	0.3509 μs	1.00	0.00	30.2734	30.2734	30.2734	234.95 KB	1.00
LodestarAdfAutolag	2000	712.069 μs	45.5692 μs	2.4978 μs	8.26	0.04	249.0234	249.0234	249.0234	1001.27 KB	4.26
CortexAugmentedDickeyFuller	2000	229.283 μs	5.1119 μs	0.2802 μs	2.66	0.01	30.2734	30.2734	30.2734	142.76 KB	0.61
LodestarKpss	2000	61.617 μs	3.7231 μs	0.2041 μs	0.71	0.00	0.1221	-	-	15.75 KB	0.07
CortexKpss	2000	62.911 μs	1.8635 μs	0.1021 μs	0.73	0.00	0.3662	-	-	31.55 KB	0.13
LodestarDecompose	2000	13.185 μs	4.1048 μs	0.2250 μs	0.15	0.00	0.7629	0.0610	-	62.75 KB	0.27
CortexDecompose	2000	26.997 μs	4.2490 μs	0.2329 μs	0.31	0.00	0.9460	0.0305	-	78.48 KB	0.33

Lodestar.Stats.Benchmarks.StatsBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleSize	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
LodestarWelchT	100	669.87 ns	19.511 ns	1.069 ns	1.00	0.00	-	-	-	-	NA
AccordWelchT	100	23,629.59 ns	603.202 ns	33.064 ns	35.27	0.06	-	-	-	392 B	NA
LodestarMannWhitney	100	1,986.22 ns	96.090 ns	5.267 ns	2.97	0.01	-	-	-	-	NA
AccordMannWhitney	100	20,667.85 ns	999.370 ns	54.779 ns	30.85	0.08	0.2747	-	-	23336 B	NA
LodestarChiSquare	100	87.26 ns	19.402 ns	1.063 ns	0.13	0.00	0.0019	-	-	168 B	NA
AccordChiSquare	100	153.46 ns	19.933 ns	1.093 ns	0.23	0.00	0.0019	-	-	168 B	NA

LodestarWelchT	10000	22,932.79 ns	4,125.021 ns	226.106 ns	1.000	0.01	-	-	-	-	NA
AccordWelchT	10000	83,731.61 ns	10,688.338 ns	585.864 ns	3.651	0.04	-	-	-	392 B	NA
LodestarMannWhitney	10000	446,946.17 ns	10,482.870 ns	574.601 ns	19.491	0.17	-	-	-	-	NA
AccordMannWhitney	10000	11,363,433.51 ns	1,068,112.602 ns	58,546.856 ns	495.542	4.78	234.3750	234.3750	234.3750	2241217 B	NA
LodestarChiSquare	10000	87.54 ns	5.840 ns	0.320 ns	0.004	0.00	0.0019	-	-	168 B	NA
AccordChiSquare	10000	153.72 ns	12.448 ns	0.682 ns	0.007	0.00	0.0019	-	-	168 B	NA

Lodestar.Text.Benchmarks.AddedTokenScanBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Gen0	Allocated
ChatTemplate	84.48 ms	2.469 ms	0.135 ms	333.3333	33.3 MB
ProseWithoutAddedTokens	71.52 ms	1.393 ms	0.076 ms	285.7143	28.47 MB

Lodestar.Text.Benchmarks.AgglomerativeIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Rows	Method	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_Fit	500	average	1,423.5 μs	534.03 μs	29.27 μs	1.00	0.03	248.0469	248.0469	248.0469	1022.55 KB	1.00
Aglomera_GetClustering	500	average	69,734.7 μs	4,997.22 μs	273.91 μs	49.00	0.88	625.0000	500.0000	-	59880.37 KB	58.56

Lodestar_Fit	500	complete	1,341.7 μs	109.30 μs	5.99 μs	1.00	0.01	248.0469	248.0469	248.0469	1022.4 KB	1.00
Aglomera_GetClustering	500	complete	81,868.7 μs	94,723.87 μs	5,192.14 μs	61.02	3.36	1000.0000	857.1429	-	83110.08 KB	81.29

Lodestar_Fit	500	single	777.2 μs	95.30 μs	5.22 μs	1.00	0.01	-	-	-	30.34 KB	1.00
Aglomera_GetClustering	500	single	107,686.3 μs	26,507.59 μs	1,452.97 μs	138.55	1.81	800.0000	200.0000	-	81288.37 KB	2,678.83

Lodestar_Fit	500	ward	1,545.8 μs	303.81 μs	16.65 μs	1.00	0.01	248.0469	248.0469	248.0469	1022.4 KB	1.00
Aglomera_GetClustering	500	ward	61,229.4 μs	20,942.11 μs	1,147.91 μs	39.61	0.74	750.0000	500.0000	-	64189.75 KB	62.78

Lodestar_Fit	1500	average	11,392.9 μs	1,945.69 μs	106.65 μs	1.00	0.01	500.0000	500.0000	500.0000	8925.15 KB	1.00
Aglomera_GetClustering	1500	average	1,379,678.3 μs	245,939.90 μs	13,480.80 μs	121.11	1.42	6000.0000	3000.0000	1000.0000	537131.64 KB	60.18

Lodestar_Fit	1500	complete	10,676.9 μs	795.73 μs	43.62 μs	1.00	0.01	500.0000	500.0000	500.0000	8925 KB	1.00
Aglomera_GetClustering	1500	complete	1,589,810.2 μs	388,722.45 μs	21,307.19 μs	148.90	1.81	10000.0000	5000.0000	2000.0000	747334.45 KB	83.73

Lodestar_Fit	1500	single	5,271.1 μs	526.57 μs	28.86 μs	1.00	0.01	-	-	-	89.92 KB	1.00
Aglomera_GetClustering	1500	single	2,503,266.5 μs	552,615.56 μs	30,290.72 μs	474.91	5.46	9000.0000	5000.0000	2000.0000	732665.26 KB	8,147.80

Lodestar_Fit	1500	ward	12,355.0 μs	1,101.38 μs	60.37 μs	1.00	0.01	500.0000	500.0000	500.0000	8925 KB	1.00
Aglomera_GetClustering	1500	ward	1,409,563.7 μs	40,272.34 μs	2,207.46 μs	114.09	0.51	8000.0000	5000.0000	2000.0000	573606.88 KB	64.27

Lodestar.Text.Benchmarks.BatchEmbeddingBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	CorpusSize	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Allocated	Alloc Ratio
UnitLoop	1	4.694 μs	0.9976 μs	0.0547 μs	1.00	0.01	0.0229	-	2.23 KB	1.00
EmbedBatch	1	5.035 μs	0.5388 μs	0.0295 μs	1.07	0.01	0.0305	-	2.63 KB	1.18
EmbedBatchBucketed	1	4.974 μs	2.5731 μs	0.1410 μs	1.06	0.03	0.0305	-	2.63 KB	1.18

UnitLoop	8	47.313 μs	11.6639 μs	0.6393 μs	1.00	0.02	0.3052	-	29.16 KB	1.00
EmbedBatch	8	16.720 μs	3.0364 μs	0.1664 μs	0.35	0.01	0.2441	-	22.19 KB	0.76
EmbedBatchBucketed	8	16.340 μs	2.8889 μs	0.1583 μs	0.35	0.00	0.2441	-	22.19 KB	0.76

UnitLoop	32	185.133 μs	9.9685 μs	0.5464 μs	1.00	0.00	1.2207	-	109.74 KB	1.00
EmbedBatch	32	62.328 μs	30.8909 μs	1.6932 μs	0.34	0.01	0.9766	-	82.35 KB	0.75
EmbedBatchBucketed	32	57.105 μs	1.4536 μs	0.0797 μs	0.31	0.00	0.7935	-	68.84 KB	0.63

UnitLoop	128	741.587 μs	20.8333 μs	1.1419 μs	1.00	0.00	4.8828	-	438.9 KB	1.00
EmbedBatch	128	242.561 μs	11.4175 μs	0.6258 μs	0.33	0.00	3.9063	-	328.54 KB	0.75
EmbedBatchBucketed	128	213.609 μs	6.7738 μs	0.3713 μs	0.29	0.00	3.1738	0.2441	261.02 KB	0.59

Lodestar.Text.Benchmarks.BertNormalizerBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Text	Lowercase	Mean	Error	StdDev	Gen0	Allocated
Encode	Ascii	False	28.14 ms	12.156 ms	0.666 ms	125.0000	11.7 MB
Encode	Ascii	True	30.11 ms	0.993 ms	0.054 ms	125.0000	11.7 MB
Encode	Accented	False	27.10 ms	5.908 ms	0.324 ms	125.0000	11.69 MB
Encode	Accented	True	38.36 ms	0.901 ms	0.049 ms	214.2857	20.38 MB
Encode	Cjk	False	32.11 ms	1.659 ms	0.091 ms	250.0000	21.55 MB
Encode	Cjk	True	44.79 ms	21.665 ms	1.188 ms	250.0000	24.48 MB

Lodestar.Text.Benchmarks.BkTreeBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Radius	Shape	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
LengthFilteredScan	1	clustered	114.18 ms	17.877 ms	0.980 ms	1.00	0.01	27.17 KB	1.00
TreeWithinDistance	1	clustered	68.08 ms	27.707 ms	1.519 ms	0.60	0.01	103.65 KB	3.81

LengthFilteredScan	1	uniform	122.11 ms	13.500 ms	0.740 ms	1.00	0.01	23.72 KB	1.00
TreeWithinDistance	1	uniform	63.30 ms	4.684 ms	0.257 ms	0.52	0.00	116.45 KB	4.91

LengthFilteredScan	2	clustered	169.73 ms	1.570 ms	0.086 ms	1.00	0.00	103.22 KB	1.00
TreeWithinDistance	2	clustered	283.31 ms	36.581 ms	2.005 ms	1.67	0.01	258.46 KB	2.50

LengthFilteredScan	2	uniform	174.76 ms	23.821 ms	1.306 ms	1.00	0.01	54.56 KB	1.00
TreeWithinDistance	2	uniform	239.85 ms	3.777 ms	0.207 ms	1.37	0.01	192.71 KB	3.53

LengthFilteredScan	3	clustered	212.49 ms	4.094 ms	0.224 ms	1.00	0.00	949.61 KB	1.00
TreeWithinDistance	3	clustered	379.04 ms	17.309 ms	0.949 ms	1.78	0.00	1366.35 KB	1.44

LengthFilteredScan	3	uniform	223.94 ms	147.663 ms	8.094 ms	1.00	0.04	741.28 KB	1.00
TreeWithinDistance	3	uniform	361.36 ms	62.349 ms	3.418 ms	1.62	0.05	1153.52 KB	1.56

LengthFilteredScan	4	clustered	250.97 ms	50.878 ms	2.789 ms	1.00	0.01	5113.18 KB	1.00
TreeWithinDistance	4	clustered	441.41 ms	36.776 ms	2.016 ms	1.76	0.02	7215.27 KB	1.41

LengthFilteredScan	4	uniform	253.12 ms	33.416 ms	1.832 ms	1.00	0.01	5513.99 KB	1.00
TreeWithinDistance	4	uniform	448.69 ms	177.699 ms	9.740 ms	1.77	0.04	7964.22 KB	1.44

Lodestar.Text.Benchmarks.BlockedTableBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	length	Mean	Error	StdDev	Allocated
Latin	1000	36.52 μs	0.218 μs	0.012 μs	-
Cjk	1000	49.28 μs	17.270 μs	0.947 μs	-
Latin	10000	3,488.78 μs	212.278 μs	11.636 μs	-
Cjk	10000	5,371.85 μs	82.631 μs	4.529 μs	-
Latin	65536	160,524.22 μs	103,424.070 μs	5,669.022 μs	-

Lodestar.Text.Benchmarks.Bm25Benchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
LodestarQuery	1000	1.619 μs	0.2716 μs	0.0149 μs	1.00	0.01	0.0038	-	-	424 B	1.00
LodestarQueryMultiTerm	1000	2.033 μs	0.0949 μs	0.0052 μs	1.26	0.01	0.0038	-	-	424 B	1.00
LuceneQuery	1000	2.674 μs	0.3797 μs	0.0208 μs	1.65	0.02	0.0610	-	-	5264 B	12.42
LodestarFromText	1000	6,662.116 μs	260.9418 μs	14.3031 μs	4,114.47	33.49	875.0000	859.3750	859.3750	4838298 B	11,411.08
LuceneFromText	1000	5,477.774 μs	183.9075 μs	10.0806 μs	3,383.03	27.35	15.6250	7.8125	-	1375027 B	3,242.99

LodestarQuery	20000	23.606 μs	8.0122 μs	0.4392 μs	1.00	0.02	-	-	-	424 B	1.00
LodestarQueryMultiTerm	20000	32.922 μs	3.9991 μs	0.2192 μs	1.39	0.02	-	-	-	424 B	1.00
LuceneQuery	20000	13.388 μs	1.7359 μs	0.0952 μs	0.57	0.01	0.0916	-	-	8648 B	20.40
LodestarFromText	20000	102,354.706 μs	39,268.2309 μs	2,152.4242 μs	4,337.02	105.01	600.0000	400.0000	400.0000	83748102 B	197,519.11
LuceneFromText	20000	109,561.935 μs	13,066.5118 μs	716.2196 μs	4,642.41	78.58	200.0000	-	-	22417147 B	52,870.63

Lodestar.Text.Benchmarks.BpeBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	Gen0	Allocated	Alloc Ratio
Unigram	23.55 ms	1.128 ms	0.062 ms	1.00	62.5000	5.43 MB	1.00
Bpe	71.03 ms	3.210 ms	0.176 ms	3.02	285.7143	28.47 MB	5.24

Lodestar.Text.Benchmarks.BpeScalingBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Gen0	Allocated
BpeOnOnePathologicalToken	512	15.00 μs	1.106 μs	0.061 μs	0.0916	7.48 KB
BpeOnOnePathologicalToken	1024	58.82 μs	5.633 μs	0.309 μs	0.1221	14.53 KB
BpeOnOnePathologicalToken	2048	151.57 μs	3.161 μs	0.173 μs	0.2441	28.58 KB
BpeOnOnePathologicalToken	4096	373.90 μs	110.610 μs	6.063 μs	0.4883	56.63 KB

Lodestar.Text.Benchmarks.BpeWordCacheBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  InvocationCount=1  IterationCount=3  
LaunchCount=1  UnrollFactor=1  WarmupCount=3

Method	Mean	Error	StdDev	Allocated
EncodeUnseenProse	7.954 ms	7.747 ms	0.4246 ms	1.45 MB

Lodestar.Text.Benchmarks.BucketRouteDiagnostics-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Alphabet	Mean	Error	StdDev	Allocated
DpGroup	cjk	12.827 μs	2.1294 μs	0.1167 μs	-
MyersGroup	cjk	209.638 μs	16.9866 μs	0.9311 μs	-
DpGroup	latin	7.132 μs	0.1716 μs	0.0094 μs	-
MyersGroup	latin	112.286 μs	13.2482 μs	0.7262 μs	-

Lodestar.Text.Benchmarks.ClassificationReportBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Classes	Mean	Error	StdDev	Gen0	Allocated
Report	10	373.5 ns	339.9 ns	18.63 ns	0.0186	1.54 KB
Report	1000	621,786.8 ns	61,086.9 ns	3,348.38 ns	0.9766	117.56 KB

Lodestar.Text.Benchmarks.ClassifierCurveBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
Roc	1000000	113.3 ms	13.36 ms	0.73 ms	800.0000	800.0000	800.0000	53.79 MB
PrecisionRecall	1000000	103.0 ms	14.85 ms	0.81 ms	800.0000	800.0000	800.0000	68.66 MB

Lodestar.Text.Benchmarks.ClusteringAgreementBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Clusters	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
AdjustedRandScore	100000	10	1.689 ms	0.1542 ms	0.0085 ms	-	-	-	12.06 KB
MutualInformationScore	100000	10	1.765 ms	0.5935 ms	0.0325 ms	-	-	-	12.06 KB
AdjustedMutualInformationScore	100000	10	11.874 ms	0.6473 ms	0.0355 ms	234.3750	234.3750	234.3750	1031.09 KB
AdjustedRandScore	100000	100	2.474 ms	0.0826 ms	0.0045 ms	199.2188	199.2188	199.2188	937.61 KB
MutualInformationScore	100000	100	2.572 ms	0.1314 ms	0.0072 ms	199.2188	199.2188	199.2188	937.61 KB
AdjustedMutualInformationScore	100000	100	101.790 ms	7.7430 ms	0.4244 ms	333.3333	333.3333	333.3333	1744.8 KB

Lodestar.Text.Benchmarks.DamerauLevenshteinBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Gen0	Allocated
Distance	12	657.0 ns	341.7 ns	18.73 ns	0.0095	840 B
Distance	120	44,924.1 ns	2,595.5 ns	142.27 ns	-	2128 B

Lodestar.Text.Benchmarks.DbscanDimensionBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_Fit	10000x8x8	201.94 ms	4.882 ms	0.268 ms	1.00	0.00	1333.3333	1333.3333	1333.3333	28.69 MB	1.00
NumFlat_Fit	10000x8x8	1,453.97 ms	22.979 ms	1.260 ms	7.20	0.01	8000.0000	7000.0000	7000.0000	156.3 MB	5.45

Lodestar_Fit	5000x16x8	63.96 ms	164.487 ms	9.016 ms	1.01	0.18	1285.7143	1285.7143	1285.7143	13.29 MB	1.00
NumFlat_Fit	5000x16x8	535.22 ms	68.811 ms	3.772 ms	8.49	1.10	3000.0000	3000.0000	3000.0000	59.77 MB	4.50

Lodestar.Text.Benchmarks.DbscanIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_Fit	20000x2x10	561.27 ms	68.530 ms	3.756 ms	1.00	0.01	2000.0000	2000.0000	2000.0000	112.13 MB	1.00
NumFlat_Fit	20000x2x10	4,289.74 ms	745.279 ms	40.851 ms	7.64	0.08	11000.0000	7000.0000	6000.0000	531.84 MB	4.74
Dbscan_CalculateClusters	20000x2x10	2,612.29 ms	158.412 ms	8.683 ms	4.65	0.03	11000.0000	10000.0000	9000.0000	372.12 MB	3.32

Lodestar_Fit	5000x2x5	44.66 ms	3.580 ms	0.196 ms	1.00	0.01	1272.7273	1272.7273	1272.7273	14.02 MB	1.00
NumFlat_Fit	5000x2x5	316.14 ms	47.129 ms	2.583 ms	7.08	0.06	2500.0000	2500.0000	2500.0000	66.59 MB	4.75
Dbscan_CalculateClusters	5000x2x5	184.71 ms	16.813 ms	0.922 ms	4.14	0.02	3666.6667	3666.6667	3666.6667	40.4 MB	2.88

Lodestar.Text.Benchmarks.DecompositionBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	RatioSD
TruncatedSvd_Rank20	16.84 ms	5.095 ms	0.279 ms	1.00	0.02
Nmf_Rank20	93.97 ms	25.100 ms	1.376 ms	5.58	0.11
MlNet_ProjectToPrincipalComponents_Rank20	15.70 ms	8.217 ms	0.450 ms	0.93	0.03

Lodestar.Text.Benchmarks.DecompositionWidthBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Columns	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
TruncatedSvd_Rank20	500	32,990.2 μs	1,333.0 μs	73.07 μs	3000.0000	3000.0000	3000.0000	12151.27 KB
TruncatedSvd_Transform	500	436.2 μs	104.5 μs	5.73 μs	91.3086	90.8203	90.8203	390.73 KB
Nmf_Frobenius_Rank20	500	192,641.1 μs	87,800.1 μs	4,812.62 μs	4500.0000	4500.0000	4500.0000	18143.39 KB
Nmf_KullbackLeibler_Rank20	500	228,977.9 μs	30,919.5 μs	1,694.80 μs	4000.0000	4000.0000	4000.0000	18596.11 KB
TruncatedSvd_Rank20	20000	194,674.0 μs	25,927.3 μs	1,421.16 μs	2666.6667	2666.6667	2666.6667	105766.49 KB
TruncatedSvd_Transform	20000	1,917.4 μs	243.2 μs	13.33 μs	496.0938	496.0938	496.0938	3437.86 KB
Nmf_Frobenius_Rank20	20000	732,028.8 μs	48,777.2 μs	2,673.65 μs	5000.0000	5000.0000	5000.0000	156850.7 KB
Nmf_KullbackLeibler_Rank20	20000	590,860.4 μs	37,207.7 μs	2,039.48 μs	5000.0000	5000.0000	5000.0000	157305.7 KB

Lodestar.Text.Benchmarks.DoubleMetaphoneBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Gen0	Allocated
Encode	224.0 μs	64.65 μs	3.54 μs	5.8594	488.6 KB

Lodestar.Text.Benchmarks.EmbeddingSearchBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Count	Mean	Error	StdDev	Allocated
SearchTop10	10000	531.7 μs	30.72 μs	1.68 μs	1.63 KB
SearchTop10	100000	8,243.3 μs	12,481.91 μs	684.18 μs	1.64 KB

Lodestar.Text.Benchmarks.EmbeddingSearchHelpersBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Rows	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
FromBlockNormalized	1000	627.0 μs	181.5 μs	9.95 μs	249.0234	249.0234	249.0234	1500.18 KB
MmrSelect100	1000	5,230.6 μs	279.6 μs	15.33 μs	-	-	-	21.02 KB
FromBlockNormalized	100000	60,250.5 μs	1,271.1 μs	69.67 μs	222.2222	222.2222	222.2222	150000.24 KB
MmrSelect100	100000	5,640.8 μs	423.7 μs	23.23 μs	-	-	-	21.02 KB

Lodestar.Text.Benchmarks.EncoderIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]     : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  Job-ETYCTQ : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

LaunchCount=1

Method	Job	InvocationCount	IterationCount	RunStrategy	UnrollFactor	WarmupCount	RowCount	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_OneHot	Job-ETYCTQ	1	20	Throughput	1	5	1000	204.86 μs	18.461 μs	20.519 μs	1.01	0.14	-	-	-	166.08 KB	1.00
Lodestar_Ordinal	Job-ETYCTQ	1	20	Throughput	1	5	1000	180.16 μs	18.340 μs	19.623 μs	0.89	0.12	-	-	-	17.53 KB	0.11
Lodestar_Impute	Job-ETYCTQ	1	20	Throughput	1	5	1000	251.08 μs	11.579 μs	12.870 μs	1.24	0.13	-	-	-	93.25 KB	0.56
MlNet_OneHotEncoding_Fit	Job-ETYCTQ	1	20	Throughput	1	5	1000	271.72 μs	15.742 μs	17.498 μs	1.34	0.15	-	-	-	34.19 KB	0.21
MlNet_OneHotEncoding_Read	Job-ETYCTQ	1	20	Throughput	1	5	1000	1,261.48 μs	95.598 μs	106.257 μs	6.21	0.77	-	-	-	363.73 KB	2.19
MlNet_ReplaceMissingValues_Read	Job-ETYCTQ	1	20	Throughput	1	5	1000	1,148.00 μs	66.040 μs	73.403 μs	5.65	0.63	-	-	-	342.21 KB	2.06

Lodestar_OneHot	ShortRun	Default	3	Default	16	3	1000	128.62 μs	42.840 μs	2.348 μs	1.00	0.02	49.8047	49.8047	49.8047	165.37 KB	1.00
Lodestar_Ordinal	ShortRun	Default	3	Default	16	3	1000	60.49 μs	6.314 μs	0.346 μs	0.47	0.01	0.1831	-	-	16.81 KB	0.10
Lodestar_Impute	ShortRun	Default	3	Default	16	3	1000	16.24 μs	1.274 μs	0.070 μs	0.13	0.00	1.1292	-	-	92.53 KB	0.56
MlNet_OneHotEncoding_Fit	ShortRun	Default	3	Default	16	3	1000	1,636.95 μs	4,926.166 μs	270.020 μs	12.73	1.83	23.4375	5.8594	-	1938.37 KB	11.72
MlNet_OneHotEncoding_Read	ShortRun	Default	3	Default	16	3	1000	746.99 μs	450.999 μs	24.721 μs	5.81	0.19	9.7656	3.9063	-	801.9 KB	4.85
MlNet_ReplaceMissingValues_Read	ShortRun	Default	3	Default	16	3	1000	208.54 μs	54.624 μs	2.994 μs	1.62	0.03	3.4180	2.9297	-	291.97 KB	1.77

Lodestar_OneHot	Job-ETYCTQ	1	20	Throughput	1	5	20000	3,785.59 μs	318.094 μs	312.411 μs	1.01	0.13	-	-	-	3282.98 KB	1.00
Lodestar_Ordinal	Job-ETYCTQ	1	20	Throughput	1	5	20000	3,566.71 μs	25.546 μs	29.419 μs	0.95	0.09	-	-	-	314.41 KB	0.10
Lodestar_Impute	Job-ETYCTQ	1	20	Throughput	1	5	20000	1,054.60 μs	22.598 μs	23.206 μs	0.28	0.03	-	-	-	1843.77 KB	0.56
MlNet_OneHotEncoding_Fit	Job-ETYCTQ	1	20	Throughput	1	5	20000	2,203.59 μs	54.448 μs	60.519 μs	0.59	0.06	-	-	-	33.53 KB	0.01
MlNet_OneHotEncoding_Read	Job-ETYCTQ	1	20	Throughput	1	5	20000	9,217.02 μs	1,021.322 μs	1,176.156 μs	2.45	0.39	-	-	-	442.96 KB	0.13
MlNet_ReplaceMissingValues_Read	Job-ETYCTQ	1	20	Throughput	1	5	20000	8,421.96 μs	261.279 μs	290.411 μs	2.24	0.23	-	-	-	532.95 KB	0.16

Lodestar_OneHot	ShortRun	Default	3	Default	16	3	20000	2,975.83 μs	1,430.037 μs	78.385 μs	1.00	0.03	992.1875	992.1875	992.1875	3290.94 KB	1.00
Lodestar_Ordinal	ShortRun	Default	3	Default	16	3	20000	2,669.35 μs	1,418.743 μs	77.766 μs	0.90	0.03	89.8438	89.8438	89.8438	314.36 KB	0.10
Lodestar_Impute	ShortRun	Default	3	Default	16	3	20000	535.75 μs	103.969 μs	5.699 μs	0.18	0.00	258.7891	258.7891	258.7891	1845.37 KB	0.56
MlNet_OneHotEncoding_Fit	ShortRun	Default	3	Default	16	3	20000	1,463.03 μs	2,184.970 μs	119.766 μs	0.49	0.04	11.7188	2.9297	-	986.36 KB	0.30
MlNet_OneHotEncoding_Read	ShortRun	Default	3	Default	16	3	20000	3,065.01 μs	1,533.274 μs	84.044 μs	1.03	0.03	-	-	-	575.2 KB	0.17
MlNet_ReplaceMissingValues_Read	ShortRun	Default	3	Default	16	3	20000	1,882.92 μs	644.015 μs	35.301 μs	0.63	0.02	3.9063	-	-	435.5 KB	0.13

Lodestar.Text.Benchmarks.FilteredVectorSearchBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Records	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
FilteredTop10	10000	512.2 μs	30.55 μs	1.67 μs	-	-	-	7.2 KB
Unfiltered	10000	545.7 μs	34.20 μs	1.87 μs	-	-	-	2.8 KB
HybridSelective	10000	2,174.4 μs	126.41 μs	6.93 μs	70.3125	58.5938	54.6875	2577.95 KB
HybridBroad	10000	3,353.1 μs	93.34 μs	5.12 μs	89.8438	78.1250	70.3125	3221.13 KB
FilteredTop10	100000	5,483.0 μs	4,578.65 μs	250.97 μs	-	-	-	7.2 KB
Unfiltered	100000	8,531.8 μs	25,121.76 μs	1,377.01 μs	-	-	-	2.81 KB
HybridSelective	100000	36,883.4 μs	2,635.00 μs	144.43 μs	-	-	-	23470.93 KB
HybridBroad	100000	51,698.1 μs	3,415.01 μs	187.19 μs	-	-	-	29359.82 KB

Lodestar.Text.Benchmarks.FuzzBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Ratio	74.28 ns	4.324 ns	0.237 ns	1.00	0.00	-	-	NA
PartialRatio	537.41 ns	6.182 ns	0.339 ns	7.24	0.02	-	-	NA
TokenSortRatio	696.30 ns	472.670 ns	25.909 ns	9.37	0.30	0.0134	1120 B	NA
TokenSetRatio	752.29 ns	19.381 ns	1.062 ns	10.13	0.03	0.0105	896 B	NA
WRatio	980.93 ns	71.854 ns	3.939 ns	13.21	0.06	0.0134	1120 B	NA

Lodestar.Text.Benchmarks.FuzzCodePointBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Allocated	Alloc Ratio
Ratio_RepeatedEmoji	1,638,686.6260 ns	74,445.6488 ns	4,080.6172 ns	1.000	0.00	-	-	160506 B	1.00
TokenSortRatio_EmojiWords	558,978.7793 ns	29,150.9202 ns	1,597.8603 ns	0.341	0.00	6.8359	1.9531	602665 B	3.75
WRatio_ProseAndOneEmoji	1,000,000.1842 ns	15,620.1919 ns	856.1954 ns	0.610	0.00	7.8125	-	720569 B	4.49
Ratio_DistinctAstral	75,915,267.6190 ns	45,152,102.3908 ns	2,474,939.1010 ns	46.327	1.31	-	-	510817 B	3.18
WRatio_EmptyOperand	0.0000 ns	0.0000 ns	0.0000 ns	0.000	0.00	-	-	-	0.00

Lodestar.Text.Benchmarks.FuzzIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Operation	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Lodestar	Ratio	73.05 ns	5.454 ns	0.299 ns	1.00	0.01	-	-	NA
FuzzySharp	Ratio	167.90 ns	139.829 ns	7.664 ns	2.30	0.09	0.0010	80 B	NA

Lodestar	PartialRatio	528.82 ns	18.676 ns	1.024 ns	1.00	0.00	-	-	NA
FuzzySharp	PartialRatio	7,555.49 ns	3,335.313 ns	182.820 ns	14.29	0.30	-	160 B	NA

Lodestar	TokenSetRatio	777.91 ns	52.984 ns	2.904 ns	1.00	0.00	0.0105	896 B	1.00
FuzzySharp	TokenSetRatio	1,498.63 ns	199.952 ns	10.960 ns	1.93	0.01	0.0229	1944 B	2.17

Lodestar	WRatio	935.20 ns	6.059 ns	0.332 ns	1.00	0.00	0.0134	1120 B	1.00
FuzzySharp	WRatio	3,992.06 ns	2,761.825 ns	151.385 ns	4.27	0.14	0.0305	3096 B	2.76

Lodestar.Text.Benchmarks.HouseholderQrBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Rows	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
Householder	2000	1.351 ms	0.0625 ms	0.0034 ms	253.9063	253.9063	250.0000	1.38 MB
Householder	20000	20.482 ms	0.3915 ms	0.0215 ms	968.7500	968.7500	968.7500	13.74 MB

Lodestar.Text.Benchmarks.IndelBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_Utf16	8	20.44 ns	3.451 ns	0.189 ns	1.00	0.01	-	NA
Distance_CodePoint	8	26.93 ns	1.963 ns	0.108 ns	1.32	0.01	-	NA
NormalizedSimilarity_Utf16	8	20.39 ns	1.214 ns	0.067 ns	1.00	0.01	-	NA
SubsequenceLength_Utf16	8	19.24 ns	2.103 ns	0.115 ns	0.94	0.01	-	NA

Distance_Utf16	12	23.28 ns	4.042 ns	0.222 ns	1.00	0.01	-	NA
Distance_CodePoint	12	29.46 ns	7.039 ns	0.386 ns	1.27	0.02	-	NA
NormalizedSimilarity_Utf16	12	23.24 ns	8.931 ns	0.490 ns	1.00	0.02	-	NA
SubsequenceLength_Utf16	12	20.24 ns	3.407 ns	0.187 ns	0.87	0.01	-	NA

Distance_Utf16	16	24.18 ns	11.463 ns	0.628 ns	1.00	0.03	-	NA
Distance_CodePoint	16	33.30 ns	2.257 ns	0.124 ns	1.38	0.03	-	NA
NormalizedSimilarity_Utf16	16	24.67 ns	3.278 ns	0.180 ns	1.02	0.02	-	NA
SubsequenceLength_Utf16	16	21.91 ns	2.040 ns	0.112 ns	0.91	0.02	-	NA

Distance_Utf16	20	26.60 ns	3.064 ns	0.168 ns	1.00	0.01	-	NA
Distance_CodePoint	20	33.15 ns	10.099 ns	0.554 ns	1.25	0.02	-	NA
NormalizedSimilarity_Utf16	20	29.37 ns	2.635 ns	0.144 ns	1.10	0.01	-	NA
SubsequenceLength_Utf16	20	25.55 ns	0.946 ns	0.052 ns	0.96	0.01	-	NA

Distance_Utf16	24	40.61 ns	1.691 ns	0.093 ns	1.00	0.00	-	NA
Distance_CodePoint	24	48.39 ns	6.924 ns	0.380 ns	1.19	0.01	-	NA
NormalizedSimilarity_Utf16	24	47.55 ns	16.049 ns	0.880 ns	1.17	0.02	-	NA
SubsequenceLength_Utf16	24	39.40 ns	0.446 ns	0.024 ns	0.97	0.00	-	NA

Distance_Utf16	32	52.31 ns	18.774 ns	1.029 ns	1.00	0.02	-	NA
Distance_CodePoint	32	53.41 ns	9.025 ns	0.495 ns	1.02	0.02	-	NA
NormalizedSimilarity_Utf16	32	51.67 ns	1.527 ns	0.084 ns	0.99	0.02	-	NA
SubsequenceLength_Utf16	32	46.59 ns	1.969 ns	0.108 ns	0.89	0.02	-	NA

Distance_Utf16	128	299.51 ns	316.545 ns	17.351 ns	1.00	0.07	-	NA
Distance_CodePoint	128	320.23 ns	29.268 ns	1.604 ns	1.07	0.05	-	NA
NormalizedSimilarity_Utf16	128	286.03 ns	34.755 ns	1.905 ns	0.96	0.05	-	NA
SubsequenceLength_Utf16	128	289.43 ns	142.338 ns	7.802 ns	0.97	0.05	-	NA

Distance_Utf16	512	3,729.13 ns	1,182.494 ns	64.817 ns	1.00	0.02	-	NA
Distance_CodePoint	512	4,006.56 ns	3,109.706 ns	170.453 ns	1.07	0.04	-	NA
NormalizedSimilarity_Utf16	512	3,768.48 ns	746.025 ns	40.892 ns	1.01	0.02	-	NA
SubsequenceLength_Utf16	512	3,865.77 ns	285.996 ns	15.676 ns	1.04	0.02	-	NA

Lodestar.Text.Benchmarks.IndelCodePointBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_CodePoint	20	303.4 ns	27.03 ns	1.48 ns	1.00	0.01	-	NA
Distance_Utf16	20	137.8 ns	37.96 ns	2.08 ns	0.45	0.01	-	NA

Distance_CodePoint	128	1,734.4 ns	201.30 ns	11.03 ns	1.00	0.01	-	NA
Distance_Utf16	128	3,066.2 ns	377.60 ns	20.70 ns	1.77	0.01	-	NA

Distance_CodePoint	512	11,190.9 ns	477.34 ns	26.16 ns	1.00	0.00	-	NA
Distance_Utf16	512	28,828.4 ns	9,048.22 ns	495.96 ns	2.58	0.04	-	NA

Lodestar.Text.Benchmarks.KMeansFitIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Lodestar_Fit	10000x16x16	5,512.8 μs	714.97 μs	39.19 μs	1.00	0.01	-	162.63 KB	1.00
NumFlat_Fit	10000x16x16	29,266.3 μs	3,648.12 μs	199.97 μs	5.31	0.05	-	10.33 KB	0.06
MetaNumerics_Fit	10000x16x16	46,705.0 μs	12,312.66 μs	674.90 μs	8.47	0.12	-	41.83 KB	0.26

Lodestar_Fit	10000x2x8	640.6 μs	43.98 μs	2.41 μs	1.00	0.00	0.9766	156.73 KB	1.00
NumFlat_Fit	10000x2x8	3,352.8 μs	489.45 μs	26.83 μs	5.23	0.04	-	3 KB	0.02
MetaNumerics_Fit	10000x2x8	2,460.9 μs	189.50 μs	10.39 μs	3.84	0.02	-	39.57 KB	0.25

Lodestar_Fit	50000x8x32	28,716.3 μs	8,452.93 μs	463.33 μs	1.00	0.02	-	787.78 KB	1.00
NumFlat_Fit	50000x8x32	325,741.0 μs	9,311.96 μs	510.42 μs	11.35	0.16	-	15.46 KB	0.02
MetaNumerics_Fit	50000x8x32	403,261.0 μs	270,418.93 μs	14,822.57 μs	14.05	0.49	-	199.91 KB	0.25

Lodestar.Text.Benchmarks.KMeansLloydIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Lodestar_Lloyd	10000x16x16	7.276 ms	0.9735 ms	0.0534 ms	1.00	0.01	88.7 KB	1.00
NumFlat_Lloyd	10000x16x16	12.205 ms	0.1826 ms	0.0100 ms	1.68	0.01	13.58 KB	0.15

Lodestar_Lloyd	10000x2x8	26.347 ms	3.1185 ms	0.1709 ms	1.00	0.01	98.92 KB	1.00
NumFlat_Lloyd	10000x2x8	78.324 ms	4.9309 ms	0.2703 ms	2.97	0.02	96.42 KB	0.97

Lodestar_Lloyd	50000x8x32	70.025 ms	1.8664 ms	0.1023 ms	1.00	0.00	410.26 KB	1.00
NumFlat_Lloyd	50000x8x32	157.674 ms	82.7912 ms	4.5381 ms	2.25	0.06	36.4 KB	0.09

Lodestar.Text.Benchmarks.LcsGateBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Band	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Dp	8	104.08 ns	4.578 ns	0.251 ns	1.00	0.00	-	NA
Kernel	8	44.63 ns	0.383 ns	0.021 ns	0.43	0.00	-	NA
Dp_Cjk	8	103.28 ns	6.785 ns	0.372 ns	0.99	0.00	-	NA
Kernel_Cjk	8	69.12 ns	0.284 ns	0.016 ns	0.66	0.00	-	NA

Dp	12	183.06 ns	3.872 ns	0.212 ns	1.00	0.00	-	NA
Kernel	12	53.88 ns	20.134 ns	1.104 ns	0.29	0.01	-	NA
Dp_Cjk	12	185.19 ns	14.403 ns	0.789 ns	1.01	0.00	-	NA
Kernel_Cjk	12	86.38 ns	43.108 ns	2.363 ns	0.47	0.01	-	NA

Dp	14	237.25 ns	20.316 ns	1.114 ns	1.00	0.01	-	NA
Kernel	14	57.06 ns	9.766 ns	0.535 ns	0.24	0.00	-	NA
Dp_Cjk	14	240.19 ns	71.797 ns	3.935 ns	1.01	0.01	-	NA
Kernel_Cjk	14	87.09 ns	10.456 ns	0.573 ns	0.37	0.00	-	NA

Dp	16	288.43 ns	54.697 ns	2.998 ns	1.00	0.01	-	NA
Kernel	16	59.37 ns	1.930 ns	0.106 ns	0.21	0.00	-	NA
Dp_Cjk	16	289.56 ns	72.643 ns	3.982 ns	1.00	0.02	-	NA
Kernel_Cjk	16	92.99 ns	2.563 ns	0.140 ns	0.32	0.00	-	NA

Dp	18	350.45 ns	19.932 ns	1.093 ns	1.00	0.00	-	NA
Kernel	18	63.20 ns	2.916 ns	0.160 ns	0.18	0.00	-	NA
Dp_Cjk	18	347.21 ns	7.915 ns	0.434 ns	0.99	0.00	-	NA
Kernel_Cjk	18	96.41 ns	5.821 ns	0.319 ns	0.28	0.00	-	NA

Dp	20	412.06 ns	23.617 ns	1.295 ns	1.00	0.00	-	NA
Kernel	20	64.15 ns	11.419 ns	0.626 ns	0.16	0.00	-	NA
Dp_Cjk	20	412.54 ns	5.339 ns	0.293 ns	1.00	0.00	-	NA
Kernel_Cjk	20	96.91 ns	1.352 ns	0.074 ns	0.24	0.00	-	NA

Dp	24	557.62 ns	37.507 ns	2.056 ns	1.00	0.00	-	NA
Kernel	24	69.57 ns	1.270 ns	0.070 ns	0.12	0.00	-	NA
Dp_Cjk	24	559.62 ns	92.237 ns	5.056 ns	1.00	0.01	-	NA
Kernel_Cjk	24	112.04 ns	5.753 ns	0.315 ns	0.20	0.00	-	NA

Dp	32	949.12 ns	156.118 ns	8.557 ns	1.00	0.01	-	NA
Kernel	32	85.15 ns	37.042 ns	2.030 ns	0.09	0.00	-	NA
Dp_Cjk	32	964.24 ns	237.367 ns	13.011 ns	1.02	0.01	-	NA
Kernel_Cjk	32	126.27 ns	7.517 ns	0.412 ns	0.13	0.00	-	NA

Dp	48	2,304.90 ns	1,345.350 ns	73.743 ns	1.00	0.04	-	NA
Kernel	48	99.42 ns	3.053 ns	0.167 ns	0.04	0.00	-	NA
Dp_Cjk	48	2,140.79 ns	1,129.607 ns	61.918 ns	0.93	0.03	-	NA
Kernel_Cjk	48	180.40 ns	37.380 ns	2.049 ns	0.08	0.00	-	NA

Dp	64	3,671.30 ns	610.884 ns	33.485 ns	1.00	0.01	-	NA
Kernel	64	121.09 ns	10.565 ns	0.579 ns	0.03	0.00	-	NA
Dp_Cjk	64	3,648.97 ns	221.184 ns	12.124 ns	0.99	0.01	-	NA
Kernel_Cjk	64	219.09 ns	16.389 ns	0.898 ns	0.06	0.00	-	NA

Dp	96	8,296.05 ns	6,329.136 ns	346.921 ns	1.00	0.05	-	NA
Kernel	96	253.66 ns	32.597 ns	1.787 ns	0.03	0.00	-	NA
Dp_Cjk	96	8,050.51 ns	516.316 ns	28.301 ns	0.97	0.03	-	NA
Kernel_Cjk	96	663.21 ns	27.203 ns	1.491 ns	0.08	0.00	-	NA

Lodestar.Text.Benchmarks.LevenshteinBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_Utf16	8	20.65 ns	4.566 ns	0.250 ns	1.00	0.01	-	NA
Distance_Utf16_Cjk	8	21.17 ns	0.282 ns	0.015 ns	1.03	0.01	-	NA
Distance_CodePoint	8	111.92 ns	3.200 ns	0.175 ns	5.42	0.06	-	NA
NormalizedSimilarity_Utf16	8	19.02 ns	2.581 ns	0.141 ns	0.92	0.01	-	NA

Distance_Utf16	64	231.42 ns	6.258 ns	0.343 ns	1.00	0.00	-	NA
Distance_Utf16_Cjk	64	291.74 ns	59.206 ns	3.245 ns	1.26	0.01	-	NA
Distance_CodePoint	64	554.47 ns	41.779 ns	2.290 ns	2.40	0.01	-	NA
NormalizedSimilarity_Utf16	64	235.02 ns	23.100 ns	1.266 ns	1.02	0.00	-	NA

Distance_Utf16	128	707.07 ns	27.092 ns	1.485 ns	1.00	0.00	-	NA
Distance_Utf16_Cjk	128	1,367.37 ns	184.952 ns	10.138 ns	1.93	0.01	-	NA
Distance_CodePoint	128	1,316.57 ns	1,057.102 ns	57.943 ns	1.86	0.07	-	NA
NormalizedSimilarity_Utf16	128	743.14 ns	212.765 ns	11.662 ns	1.05	0.01	-	NA

Distance_Utf16	512	9,795.42 ns	947.925 ns	51.959 ns	1.00	0.01	-	NA
Distance_Utf16_Cjk	512	13,972.56 ns	2,390.658 ns	131.040 ns	1.43	0.01	-	NA
Distance_CodePoint	512	11,175.80 ns	429.426 ns	23.538 ns	1.14	0.01	-	NA
NormalizedSimilarity_Utf16	512	9,800.24 ns	3,082.810 ns	168.979 ns	1.00	0.02	-	NA

Lodestar.Text.Benchmarks.LevenshteinCodePointBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Distinct	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_CodePoint	16	32	278.6 ns	11.15 ns	0.61 ns	1.00	0.00	-	NA
Distance_Utf16	16	32	181.9 ns	70.32 ns	3.85 ns	0.65	0.01	-	NA

Distance_CodePoint	16	512	272.3 ns	22.39 ns	1.23 ns	1.00	0.01	-	NA
Distance_Utf16	16	512	182.9 ns	59.67 ns	3.27 ns	0.67	0.01	-	NA

Distance_CodePoint	24	32	354.4 ns	65.49 ns	3.59 ns	1.00	0.01	-	NA
Distance_Utf16	24	32	255.7 ns	27.87 ns	1.53 ns	0.72	0.01	-	NA

Distance_CodePoint	24	512	357.0 ns	27.37 ns	1.50 ns	1.00	0.01	-	NA
Distance_Utf16	24	512	264.1 ns	78.35 ns	4.29 ns	0.74	0.01	-	NA

Distance_CodePoint	32	32	434.8 ns	18.11 ns	0.99 ns	1.00	0.00	-	NA
Distance_Utf16	32	32	326.1 ns	4.81 ns	0.26 ns	0.75	0.00	-	NA

Distance_CodePoint	32	512	428.0 ns	61.29 ns	3.36 ns	1.00	0.01	-	NA
Distance_Utf16	32	512	328.4 ns	25.83 ns	1.42 ns	0.77	0.01	-	NA

Distance_CodePoint	40	32	512.6 ns	110.58 ns	6.06 ns	1.00	0.01	-	NA
Distance_Utf16	40	32	918.9 ns	49.95 ns	2.74 ns	1.79	0.02	-	NA

Distance_CodePoint	40	512	495.2 ns	3.15 ns	0.17 ns	1.00	0.00	-	NA
Distance_Utf16	40	512	897.0 ns	32.12 ns	1.76 ns	1.81	0.00	-	NA

Distance_CodePoint	128	32	1,535.5 ns	62.03 ns	3.40 ns	1.00	0.00	-	NA
Distance_Utf16	128	32	4,166.0 ns	29.35 ns	1.61 ns	2.71	0.01	-	NA

Distance_CodePoint	128	512	1,581.2 ns	89.45 ns	4.90 ns	1.00	0.00	-	NA
Distance_Utf16	128	512	4,084.9 ns	2,445.96 ns	134.07 ns	2.58	0.07	-	NA

Distance_CodePoint	512	32	12,441.5 ns	743.65 ns	40.76 ns	1.00	0.00	-	NA
Distance_Utf16	512	32	51,215.7 ns	8,731.15 ns	478.58 ns	4.12	0.04	-	NA

Distance_CodePoint	512	512	445,181.5 ns	119,186.22 ns	6,533.00 ns	1.00	0.02	-	NA
Distance_Utf16	512	512	51,019.3 ns	4,196.02 ns	230.00 ns	0.11	0.00	-	NA

Lodestar.Text.Benchmarks.LevenshteinIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Lodestar	8	20.85 ns	0.433 ns	0.024 ns	1.00	0.00	-	-	NA
Fastenshtein	8	79.09 ns	6.608 ns	0.362 ns	3.79	0.02	0.0006	56 B	NA
Quickenshtein	8	67.82 ns	71.344 ns	3.911 ns	3.25	0.16	-	-	NA
F23_StringSimilarity	8	161.21 ns	53.864 ns	2.952 ns	7.73	0.12	0.0014	128 B	NA

Lodestar	64	231.40 ns	7.760 ns	0.425 ns	1.00	0.00	-	-	NA
Fastenshtein	64	4,560.13 ns	2,729.584 ns	149.618 ns	19.71	0.56	-	280 B	NA
Quickenshtein	64	1,236.87 ns	599.495 ns	32.860 ns	5.35	0.12	-	-	NA
F23_StringSimilarity	64	6,944.77 ns	969.780 ns	53.157 ns	30.01	0.20	-	576 B	NA

Lodestar	512	9,313.50 ns	172.475 ns	9.454 ns	1.00	0.00	-	-	NA
Fastenshtein	512	380,748.71 ns	19,731.458 ns	1,081.548 ns	40.88	0.11	-	2072 B	NA
Quickenshtein	512	34,051.35 ns	1,712.960 ns	93.893 ns	3.66	0.01	-	-	NA
F23_StringSimilarity	512	649,192.94 ns	136,035.397 ns	7,456.559 ns	69.70	0.70	-	4161 B	NA

Lodestar.Text.Benchmarks.MetaNumericsPcaBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_ExplainedVarianceRatio	2000x10	123.32 μs	13.901 μs	0.762 μs	1.00	0.01	-	-	-	2.38 KB	1.00
MetaNumerics_VarianceFraction	2000x10	103,493.96 μs	18,023.447 μs	987.926 μs	839.23	8.26	800.0000	800.0000	800.0000	31412.86 KB	13,226.47

Lodestar_ExplainedVarianceRatio	2000x50	2,491.54 μs	235.623 μs	12.915 μs	1.00	0.01	-	-	-	42.07 KB	1.00
MetaNumerics_VarianceFraction	2000x50	508,021.51 μs	34,138.348 μs	1,871.238 μs	203.90	1.12	-	-	-	32054.48 KB	762.01

Lodestar_ExplainedVarianceRatio	200x10	19.37 μs	5.265 μs	0.289 μs	1.00	0.02	-	-	-	2.38 KB	1.00
MetaNumerics_VarianceFraction	200x10	654.19 μs	249.849 μs	13.695 μs	33.78	0.75	99.6094	99.6094	99.6094	329.71 KB	138.82

Lodestar.Text.Benchmarks.MetricsBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Classes	Mean	Error	StdDev	Gen0	Allocated
Matrix	1000	2	3,088.2 ns	850.69 ns	46.63 ns	0.0038	376 B
MatrixWeighted	1000	2	5,056.7 ns	409.70 ns	22.46 ns	-	376 B
AccuracyScore	1000	2	113.8 ns	1.73 ns	0.09 ns	-	-
F1Macro	1000	2	2,876.8 ns	2,465.72 ns	135.15 ns	0.0038	536 B
Report	1000	2	5,011.3 ns	819.47 ns	44.92 ns	0.0610	5344 B
Matrix	1000	10	3,135.3 ns	168.50 ns	9.24 ns	0.0153	1312 B
MatrixWeighted	1000	10	5,331.2 ns	91.08 ns	4.99 ns	0.0153	1312 B
AccuracyScore	1000	10	114.3 ns	4.41 ns	0.24 ns	-	-
F1Macro	1000	10	2,959.5 ns	126.23 ns	6.92 ns	0.0191	1728 B
Report	1000	10	7,776.4 ns	359.55 ns	19.71 ns	0.1373	12336 B
Matrix	100000	2	295,770.5 ns	9,805.21 ns	537.46 ns	-	376 B
MatrixWeighted	100000	2	620,868.9 ns	147,563.59 ns	8,088.46 ns	-	377 B
AccuracyScore	100000	2	13,400.6 ns	4,590.11 ns	251.60 ns	-	-
F1Macro	100000	2	295,350.8 ns	30,760.05 ns	1,686.06 ns	-	536 B
Report	100000	2	276,604.6 ns	12,962.77 ns	710.53 ns	-	5368 B
Matrix	100000	10	252,469.4 ns	25,140.57 ns	1,378.04 ns	-	1312 B
MatrixWeighted	100000	10	746,983.1 ns	366,370.95 ns	20,082.03 ns	-	1313 B
AccuracyScore	100000	10	13,401.5 ns	800.48 ns	43.88 ns	-	-
F1Macro	100000	10	274,709.7 ns	144,847.68 ns	7,939.59 ns	-	1728 B
Report	100000	10	277,229.8 ns	84,309.79 ns	4,621.30 ns	-	12680 B
Matrix	1000000	2	2,760,355.7 ns	183,873.96 ns	10,078.75 ns	-	379 B
MatrixWeighted	1000000	2	6,442,240.8 ns	4,285,686.26 ns	234,912.93 ns	-	382 B
AccuracyScore	1000000	2	217,662.4 ns	18,099.05 ns	992.07 ns	-	-
F1Macro	1000000	2	3,000,681.3 ns	1,271,612.67 ns	69,701.38 ns	-	539 B
Report	1000000	2	2,821,188.6 ns	1,644,781.70 ns	90,156.04 ns	-	5387 B
Matrix	1000000	10	2,519,368.3 ns	180,674.02 ns	9,903.35 ns	-	1315 B
MatrixWeighted	1000000	10	7,577,830.0 ns	655,101.14 ns	35,908.30 ns	-	1318 B
AccuracyScore	1000000	10	220,762.8 ns	1,304.41 ns	71.50 ns	-	-
F1Macro	1000000	10	2,721,907.4 ns	1,338,409.81 ns	73,362.76 ns	-	1731 B
Report	1000000	10	2,669,688.0 ns	159,102.10 ns	8,720.92 ns	-	12723 B

Lodestar.Text.Benchmarks.MetricsIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Request	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar	100000	Bundle	7,413.73 μs	30.246 μs	1.658 μs	1.00	0.00	-	-	-	1064 B	1.00
MlNet	100000	Bundle	30,006.37 μs	1,107.383 μs	60.699 μs	4.05	0.01	593.7500	593.7500	593.7500	5089912 B	4,783.75

Lodestar	100000	AccuracyAlone	13.39 μs	4.032 μs	0.221 μs	1.00	0.02	-	-	-	-	NA
MlNet	100000	AccuracyAlone	30,105.78 μs	9,123.647 μs	500.098 μs	2,249.26	45.41	593.7500	593.7500	593.7500	5089895 B	NA

Lodestar	1000000	Bundle	141,907.52 μs	74,292.792 μs	4,072.239 μs	1.00	0.04	-	-	-	-	NA
MlNet	1000000	Bundle	209,724.91 μs	9,516.196 μs	521.615 μs	1.48	0.04	-	-	-	23229116 B	NA

Lodestar	1000000	AccuracyAlone	221.49 μs	26.823 μs	1.470 μs	1.00	0.01	-	-	-	-	NA
MlNet	1000000	AccuracyAlone	184,425.45 μs	14,660.434 μs	803.588 μs	832.69	5.72	-	-	-	23228972 B	NA

Lodestar.Text.Benchmarks.MultiClassRocAucBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Mean	Error	StdDev	Allocated
OneVsRest	100000	36.02 ms	1.731 ms	0.095 ms	-
OneVsOne	100000	75.23 ms	15.062 ms	0.826 ms	801865 B
OneVsRest	1000000	678.58 ms	65.703 ms	3.601 ms	-
OneVsOne	1000000	1,127.26 ms	314.592 ms	17.244 ms	8003648 B

Lodestar.Text.Benchmarks.MultilabelConfusionMatrixBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Rows	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
PerLabel	100000	6.887 ms	2.2742 ms	0.1247 ms	-	-	-	6.28 KB
PerLabelWeighted	100000	9.484 ms	1.4048 ms	0.0770 ms	-	-	-	6.29 KB
PerSample	100000	44.840 ms	44.5859 ms	2.4439 ms	437.5000	375.0000	125.0000	31250.15 KB
PerClass	100000	3.813 ms	0.0259 ms	0.0014 ms	-	-	-	6.42 KB

Lodestar.Text.Benchmarks.MyersGateBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Band	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Dp	4	59.97 ns	3.925 ns	0.215 ns	1.00	0.00	-	NA
Kernel	4	60.26 ns	1.970 ns	0.108 ns	1.00	0.00	-	NA
Dp_Cjk	4	60.90 ns	14.283 ns	0.783 ns	1.02	0.01	-	NA
Kernel_Cjk	4	61.40 ns	40.327 ns	2.210 ns	1.02	0.03	-	NA

Dp	6	81.37 ns	16.482 ns	0.903 ns	1.00	0.01	-	NA
Kernel	6	56.96 ns	0.733 ns	0.040 ns	0.70	0.01	-	NA
Dp_Cjk	6	81.53 ns	1.628 ns	0.089 ns	1.00	0.01	-	NA
Kernel_Cjk	6	94.45 ns	6.373 ns	0.349 ns	1.16	0.01	-	NA

Dp	8	110.98 ns	16.197 ns	0.888 ns	1.00	0.01	-	NA
Kernel	8	67.97 ns	0.959 ns	0.053 ns	0.61	0.00	-	NA
Dp_Cjk	8	110.37 ns	2.922 ns	0.160 ns	0.99	0.01	-	NA
Kernel_Cjk	8	123.59 ns	6.328 ns	0.347 ns	1.11	0.01	-	NA

Dp	10	152.71 ns	38.868 ns	2.130 ns	1.00	0.02	-	NA
Kernel	10	78.40 ns	22.327 ns	1.224 ns	0.51	0.01	-	NA
Dp_Cjk	10	150.33 ns	10.718 ns	0.587 ns	0.98	0.01	-	NA
Kernel_Cjk	10	108.52 ns	33.585 ns	1.841 ns	0.71	0.01	-	NA

Dp	12	207.60 ns	8.484 ns	0.465 ns	1.00	0.00	-	NA
Kernel	12	82.82 ns	0.732 ns	0.040 ns	0.40	0.00	-	NA
Dp_Cjk	12	206.27 ns	115.185 ns	6.314 ns	0.99	0.03	-	NA
Kernel_Cjk	12	116.21 ns	31.256 ns	1.713 ns	0.56	0.01	-	NA

Dp	16	309.16 ns	6.135 ns	0.336 ns	1.00	0.00	-	NA
Kernel	16	102.38 ns	34.811 ns	1.908 ns	0.33	0.01	-	NA
Dp_Cjk	16	323.03 ns	110.557 ns	6.060 ns	1.04	0.02	-	NA
Kernel_Cjk	16	161.46 ns	5.848 ns	0.321 ns	0.52	0.00	-	NA

Dp	24	673.13 ns	666.399 ns	36.528 ns	1.00	0.07	-	NA
Kernel	24	131.60 ns	58.237 ns	3.192 ns	0.20	0.01	-	NA
Dp_Cjk	24	640.21 ns	44.420 ns	2.435 ns	0.95	0.04	-	NA
Kernel_Cjk	24	171.69 ns	15.246 ns	0.836 ns	0.26	0.01	-	NA

Dp	32	1,135.57 ns	79.480 ns	4.357 ns	1.00	0.00	-	NA
Kernel	32	160.34 ns	5.947 ns	0.326 ns	0.14	0.00	-	NA
Dp_Cjk	32	1,139.77 ns	71.823 ns	3.937 ns	1.00	0.00	-	NA
Kernel_Cjk	32	216.12 ns	9.455 ns	0.518 ns	0.19	0.00	-	NA

Dp	48	2,509.83 ns	1,075.375 ns	58.945 ns	1.00	0.03	-	NA
Kernel	48	227.36 ns	101.823 ns	5.581 ns	0.09	0.00	-	NA
Dp_Cjk	48	2,482.35 ns	300.676 ns	16.481 ns	0.99	0.02	-	NA
Kernel_Cjk	48	305.58 ns	384.847 ns	21.095 ns	0.12	0.01	-	NA

Dp	64	4,623.16 ns	782.302 ns	42.881 ns	1.00	0.01	-	NA
Kernel	64	286.39 ns	3.365 ns	0.184 ns	0.06	0.00	-	NA
Dp_Cjk	64	4,474.60 ns	678.537 ns	37.193 ns	0.97	0.01	-	NA
Kernel_Cjk	64	358.01 ns	12.964 ns	0.711 ns	0.08	0.00	-	NA

Dp	96	9,755.91 ns	160.477 ns	8.796 ns	1.00	0.00	-	NA
Kernel	96	594.54 ns	43.305 ns	2.374 ns	0.06	0.00	-	NA
Dp_Cjk	96	9,787.17 ns	116.758 ns	6.400 ns	1.00	0.00	-	NA
Kernel_Cjk	96	1,162.75 ns	86.087 ns	4.719 ns	0.12	0.00	-	NA

Lodestar.Text.Benchmarks.OsaBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Distance_Utf16	8	24.55 ns	10.792 ns	0.592 ns	1.00	0.03	-	NA
Distance_CodePoint	8	121.27 ns	25.220 ns	1.382 ns	4.94	0.11	-	NA

Distance_Utf16	32	117.09 ns	19.747 ns	1.082 ns	1.00	0.01	-	NA
Distance_CodePoint	32	1,037.09 ns	40.758 ns	2.234 ns	8.86	0.07	-	NA

Distance_Utf16	64	261.79 ns	6.934 ns	0.380 ns	1.00	0.00	-	NA
Distance_CodePoint	64	5,979.51 ns	786.793 ns	43.127 ns	22.84	0.15	-	NA

Distance_Utf16	128	31,929.73 ns	1,935.726 ns	106.104 ns	1.00	0.00	-	NA
Distance_CodePoint	128	32,193.02 ns	2,436.007 ns	133.526 ns	1.01	0.00	-	NA

Lodestar.Text.Benchmarks.PartialFitBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	RowCount	BatchCount	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
WholeFit	10000	10	331.2 μs	125.93 μs	6.90 μs	1.00	0.03	808 B	1.00
BatchedFit	10000	10	421.1 μs	21.28 μs	1.17 μs	1.27	0.02	3680 B	4.55
SparseFit	10000	10	119.4 μs	36.08 μs	1.98 μs	0.36	0.01	704 B	0.87
DenseFitOfTheSameData	10000	10	414.0 μs	128.61 μs	7.05 μs	1.25	0.03	784 B	0.97

WholeFit	10000	100	291.0 μs	17.24 μs	0.94 μs	1.00	0.00	808 B	1.00
BatchedFit	10000	100	559.1 μs	60.02 μs	3.29 μs	1.92	0.01	36801 B	45.55
SparseFit	10000	100	120.5 μs	7.54 μs	0.41 μs	0.41	0.00	704 B	0.87
DenseFitOfTheSameData	10000	100	418.2 μs	130.14 μs	7.13 μs	1.44	0.02	784 B	0.97

WholeFit	100000	10	3,039.1 μs	106.26 μs	5.82 μs	1.00	0.00	811 B	1.00
BatchedFit	100000	10	4,054.3 μs	212.72 μs	11.66 μs	1.33	0.00	3686 B	4.55
SparseFit	100000	10	1,664.2 μs	151.37 μs	8.30 μs	0.55	0.00	705 B	0.87
DenseFitOfTheSameData	100000	10	4,561.7 μs	13,685.20 μs	750.13 μs	1.50	0.21	790 B	0.97

WholeFit	100000	100	3,255.2 μs	596.35 μs	32.69 μs	1.00	0.01	811 B	1.00
BatchedFit	100000	100	4,640.5 μs	275.26 μs	15.09 μs	1.43	0.01	36806 B	45.38
SparseFit	100000	100	1,676.4 μs	175.91 μs	9.64 μs	0.52	0.01	705 B	0.87
DenseFitOfTheSameData	100000	100	4,000.1 μs	704.08 μs	38.59 μs	1.23	0.01	790 B	0.97

Lodestar.Text.Benchmarks.PartialRatioLongNeedleBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	NeedleLength	Mean	Error	StdDev	Allocated
Embedded	65	6.840 μs	2.7418 μs	0.1503 μs	-
EqualLength	65	2.023 μs	0.0832 μs	0.0046 μs	-
Embedded	128	25.086 μs	3.3146 μs	0.1817 μs	-
EqualLength	128	3.945 μs	1.0926 μs	0.0599 μs	-
Embedded	512	890.485 μs	100.0917 μs	5.4864 μs	-
EqualLength	512	32.976 μs	7.2542 μs	0.3976 μs	-

Lodestar.Text.Benchmarks.PartitionValidityBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
DaviesBouldinScore	1000000	7.769 ms	0.5825 ms	0.0319 ms	156.2500	156.2500	156.2500	3.82 MB
CalinskiHarabaszScore	1000000	11.010 ms	4.7985 ms	0.2630 ms	140.6250	140.6250	140.6250	3.82 MB

Lodestar.Text.Benchmarks.PersistenceBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
VocabTxt	2.748 ms	1.9967 ms	0.1094 ms	39.0625	35.1563	23.4375	3711.56 KB
TokenizerJsonWordPiece	7.494 ms	5.3543 ms	0.2935 ms	54.6875	46.8750	31.2500	5852.24 KB
TokenizerJsonUnigram	8.087 ms	2.6216 ms	0.1437 ms	-	-	-	4748.79 KB
SpieceModel	2.414 ms	1.8542 ms	0.1016 ms	35.1563	31.2500	19.5313	3440.04 KB
TfidfSave	1.435 ms	0.4300 ms	0.0236 ms	19.5313	19.5313	19.5313	2137.03 KB
TfidfLoad	2.934 ms	1.4139 ms	0.0775 ms	15.6250	7.8125	7.8125	2930.47 KB
EmbeddingIndexSave	4.834 ms	3.8278 ms	0.2098 ms	187.5000	187.5000	187.5000	20349.83 KB
EmbeddingIndexLoad	6.269 ms	0.2894 ms	0.0159 ms	125.0000	117.1875	117.1875	16094.35 KB
EmbeddingIndexSaveFile	120.718 ms	432.6402 ms	23.7145 ms	-	-	-	323.67 KB
EmbeddingIndexLoadGzip	70.104 ms	4.6278 ms	0.2537 ms	-	-	-	16095.21 KB

Lodestar.Text.Benchmarks.PrecompiledNormalizerBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Mean	Error	StdDev	Gen0	Allocated
NormalizeDocuments	11.42 ms	2.303 ms	0.126 ms	31.2500	2.75 MB
EncodeDocuments	49.62 ms	2.625 ms	0.144 ms	100.0000	8.19 MB

Lodestar.Text.Benchmarks.PrincipalComponentVarianceBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_ExplainedVariance	100x200	6,327.87 μs	240.301 μs	13.172 μs	1.00	0.00	46.8750	46.8750	46.8750	318.27 KB	1.00
NumFlat_Pca	100x200	14,090.38 μs	2,686.778 μs	147.271 μs	2.23	0.02	187.5000	187.5000	187.5000	628.54 KB	1.97

Lodestar_ExplainedVariance	2000x10	119.16 μs	8.138 μs	0.446 μs	1.00	0.00	-	-	-	2.38 KB	1.00
NumFlat_Pca	2000x10	169.47 μs	13.932 μs	0.764 μs	1.42	0.01	-	-	-	1.94 KB	0.82

Lodestar_ExplainedVariance	2000x50	2,577.10 μs	1,087.746 μs	59.623 μs	1.00	0.03	-	-	-	42.07 KB	1.00
NumFlat_Pca	2000x50	2,790.48 μs	82.736 μs	4.535 μs	1.08	0.02	-	-	-	40.07 KB	0.95

Lodestar_ExplainedVariance	200x10	19.43 μs	4.740 μs	0.260 μs	1.00	0.02	-	-	-	2.38 KB	1.00
NumFlat_Pca	200x10	23.22 μs	1.957 μs	0.107 μs	1.20	0.01	-	-	-	1.94 KB	0.82

Lodestar.Text.Benchmarks.ProcessExtractBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Limit	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Extract	1	5.650 ms	1.5889 ms	0.0871 ms	1.00	0.02	118 B	1.00
ExtractOne	1	5.785 ms	2.2498 ms	0.1233 ms	1.02	0.02	-	0.00

Extract	5	5.560 ms	0.3533 ms	0.0194 ms	1.00	0.00	214 B	1.00
ExtractOne	5	5.785 ms	0.3755 ms	0.0206 ms	1.04	0.00	-	0.00

Lodestar.Text.Benchmarks.QgramBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Q	Mean	Error	StdDev	Allocated
JaccardSimilarity	1	3.585 μs	0.5498 μs	0.0301 μs	192 B
JaccardSimilarity	3	4.459 μs	0.0531 μs	0.0029 μs	192 B

Lodestar.Text.Benchmarks.RankingMetricsBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Rows	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
NdcgTieAveraged	100000	27.242 ms	6.6191 ms	0.3628 ms	-	-	-	800310 B
NdcgIgnoringTies	100000	25.029 ms	0.7141 ms	0.0391 ms	-	-	-	800319 B
DcgTieAveraged	100000	15.862 ms	0.3850 ms	0.0211 ms	-	-	-	800215 B
ReciprocalRankScore	100000	13.428 ms	0.5142 ms	0.0282 ms	-	-	-	180 B
CoverageErrorScore	100000	6.018 ms	0.1836 ms	0.0101 ms	15.6250	15.6250	15.6250	800271 B
LabelRankingAveragePrecisionScore	100000	47.588 ms	0.7737 ms	0.0424 ms	-	-	-	-

Lodestar.Text.Benchmarks.RatcliffObershelpBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Length	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Similarity_Containment	64	311.8 ns	91.66 ns	5.02 ns	1.00	0.02	-	NA
Similarity_NearDuplicate	64	3,742.1 ns	237.49 ns	13.02 ns	12.00	0.17	-	NA

Similarity_Containment	512	15,351.4 ns	5,252.43 ns	287.90 ns	1.00	0.02	-	NA
Similarity_NearDuplicate	512	465,508.9 ns	58,435.29 ns	3,203.04 ns	30.33	0.52	-	NA

Lodestar.Text.Benchmarks.RegressionMetricsBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
Mse	100000	46.01 μs	4.365 μs	0.239 μs	-	-	-	-
Mae	100000	45.23 μs	16.874 μs	0.925 μs	-	-	-	-
R2Score	100000	116.27 μs	39.724 μs	2.177 μs	-	-	-	64 B
MedianAe	100000	465.30 μs	117.948 μs	6.465 μs	199.7070	199.7070	199.7070	800186 B
Mse	1000000	494.03 μs	51.164 μs	2.804 μs	-	-	-	-
Mae	1000000	485.73 μs	12.703 μs	0.696 μs	-	-	-	-
R2Score	1000000	1,195.21 μs	653.774 μs	35.836 μs	-	-	-	65 B
MedianAe	1000000	5,620.36 μs	576.458 μs	31.598 μs	320.3125	320.3125	320.3125	8000267 B

Lodestar.Text.Benchmarks.RobustScalerSparseBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Shape	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
Fit	2000x2000x20	21.59 ms	0.239 ms	0.013 ms	93.7500	93.7500	93.7500	375.3 KB
Fit	500x20000x5	41.07 ms	2.231 ms	0.122 ms	-	-	-	492.46 KB

Lodestar.Text.Benchmarks.ScalerIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]     : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  Job-UAQQOP : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

LaunchCount=1

Method	Job	InvocationCount	IterationCount	RunStrategy	UnrollFactor	WarmupCount	RowCount	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_MinMax	Job-UAQQOP	1	20	Throughput	1	5	1000	259.50 μs	30.922 μs	35.609 μs	1.02	0.20	-	-	-	79.59 KB	1.00
Lodestar_MaxAbs	Job-UAQQOP	1	20	Throughput	1	5	1000	285.91 μs	10.795 μs	11.998 μs	1.12	0.17	-	-	-	79.14 KB	0.99
Lodestar_Robust	Job-UAQQOP	1	20	Throughput	1	5	1000	1,759.04 μs	21.492 μs	23.888 μs	6.91	0.98	-	-	-	157.23 KB	1.98
Lodestar_Standard	Job-UAQQOP	1	20	Throughput	1	5	1000	238.04 μs	8.936 μs	9.932 μs	0.93	0.14	-	-	-	79.66 KB	1.00
MlNet_NormalizeMinMax_Fit	Job-UAQQOP	1	20	Throughput	1	5	1000	265.70 μs	10.717 μs	11.912 μs	1.04	0.15	-	-	-	8.98 KB	0.11
MlNet_NormalizeMinMax_Read	Job-UAQQOP	1	20	Throughput	1	5	1000	1,779.56 μs	50.255 μs	55.858 μs	6.99	1.01	-	-	-	324.38 KB	4.08
MlNet_NormalizeRobustScaling_Read	Job-UAQQOP	1	20	Throughput	1	5	1000	4,026.56 μs	116.577 μs	129.575 μs	15.81	2.30	-	-	-	467.07 KB	5.87

Lodestar_MinMax	ShortRun	Default	3	Default	16	3	1000	38.12 μs	22.229 μs	1.218 μs	1.00	0.04	0.9155	-	-	78.87 KB	1.00
Lodestar_MaxAbs	ShortRun	Default	3	Default	16	3	1000	35.73 μs	5.855 μs	0.321 μs	0.94	0.03	0.9155	-	-	78.4 KB	0.99
Lodestar_Robust	ShortRun	Default	3	Default	16	3	1000	415.08 μs	139.046 μs	7.622 μs	10.90	0.34	1.4648	-	-	156.79 KB	1.99
Lodestar_Standard	ShortRun	Default	3	Default	16	3	1000	54.09 μs	5.962 μs	0.327 μs	1.42	0.04	0.9155	-	-	78.71 KB	1.00
MlNet_NormalizeMinMax_Fit	ShortRun	Default	3	Default	16	3	1000	1,320.20 μs	4,430.169 μs	242.833 μs	34.66	5.60	16.6016	4.3945	-	1366.2 KB	17.32
MlNet_NormalizeMinMax_Read	ShortRun	Default	3	Default	16	3	1000	868.15 μs	1,219.680 μs	66.855 μs	22.79	1.64	7.8125	-	-	638.91 KB	8.10
MlNet_NormalizeRobustScaling_Read	ShortRun	Default	3	Default	16	3	1000	1,491.16 μs	697.729 μs	38.245 μs	39.14	1.38	3.9063	-	-	626.03 KB	7.94

Lodestar_MinMax	Job-UAQQOP	1	20	Throughput	1	5	20000	1,269.16 μs	14.165 μs	16.312 μs	1.00	0.02	-	-	-	1563.68 KB	1.000
Lodestar_MaxAbs	Job-UAQQOP	1	20	Throughput	1	5	20000	1,233.24 μs	34.276 μs	36.675 μs	0.97	0.03	-	-	-	1563.52 KB	1.000
Lodestar_Robust	Job-UAQQOP	1	20	Throughput	1	5	20000	12,998.17 μs	87.483 μs	97.238 μs	10.24	0.15	-	-	-	3126.26 KB	1.999
Lodestar_Standard	Job-UAQQOP	1	20	Throughput	1	5	20000	1,585.19 μs	33.941 μs	39.087 μs	1.25	0.03	-	-	-	1563.75 KB	1.000
MlNet_NormalizeMinMax_Fit	Job-UAQQOP	1	20	Throughput	1	5	20000	3,559.72 μs	23.113 μs	26.617 μs	2.81	0.04	-	-	-	9.3 KB	0.006
MlNet_NormalizeMinMax_Read	Job-UAQQOP	1	20	Throughput	1	5	20000	13,514.10 μs	5,278.612 μs	6,078.858 μs	10.65	4.68	-	-	-	505.91 KB	0.324
MlNet_NormalizeRobustScaling_Read	Job-UAQQOP	1	20	Throughput	1	5	20000	16,777.26 μs	819.071 μs	910.395 μs	13.22	0.72	-	-	-	4176.37 KB	2.671

Lodestar_MinMax	ShortRun	Default	3	Default	16	3	20000	1,002.78 μs	815.564 μs	44.704 μs	1.00	0.05	332.0313	332.0313	332.0313	1563.68 KB	1.00
Lodestar_MaxAbs	ShortRun	Default	3	Default	16	3	20000	896.68 μs	91.111 μs	4.994 μs	0.90	0.04	332.0313	332.0313	332.0313	1562.98 KB	1.00
Lodestar_Robust	ShortRun	Default	3	Default	16	3	20000	12,612.52 μs	4,857.859 μs	266.276 μs	12.59	0.54	890.6250	890.6250	890.6250	3126.11 KB	2.00
Lodestar_Standard	ShortRun	Default	3	Default	16	3	20000	1,269.44 μs	315.006 μs	17.267 μs	1.27	0.05	332.0313	332.0313	332.0313	1563.3 KB	1.00
MlNet_NormalizeMinMax_Fit	ShortRun	Default	3	Default	16	3	20000	1,255.20 μs	2,049.974 μs	112.366 μs	1.25	0.11	7.8125	1.9531	-	686.19 KB	0.44
MlNet_NormalizeMinMax_Read	ShortRun	Default	3	Default	16	3	20000	3,386.19 μs	3,379.568 μs	185.246 μs	3.38	0.21	-	-	-	581.16 KB	0.37
MlNet_NormalizeRobustScaling_Read	ShortRun	Default	3	Default	16	3	20000	15,849.18 μs	3,171.000 μs	173.813 μs	15.83	0.64	31.2500	-	-	4195.55 KB	2.68

Lodestar.Text.Benchmarks.SentencePieceBpeLineageBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Model	Mean	Error	StdDev	Gen0	Allocated
EncodeDocuments	Llama2	131.0 ms	73.33 ms	4.02 ms	250.0000	37.05 MB
EncodeDocuments	Mistral	122.5 ms	7.49 ms	0.41 ms	400.0000	37.03 MB

Lodestar.Text.Benchmarks.SilhouetteBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Samples	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
PerSample	2000	22.03 ms	0.708 ms	0.039 ms	31.2500	31.2500	31.2500	148.75 KB
PerSample	5000	141.00 ms	9.018 ms	0.494 ms	-	-	-	371.54 KB

Lodestar.Text.Benchmarks.SimilaritySketchBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Permutations	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
ExactPairwise	500	64	49.498 ms	4.1933 ms	0.2298 ms	1.00	0.01	-	-	-	8152.42 KB	1.00
SketchThenVerify	500	64	9.905 ms	2.4422 ms	0.1339 ms	0.20	0.00	78.1250	78.1250	78.1250	2492.48 KB	0.31
SignaturesOnly	500	64	8.439 ms	0.3915 ms	0.0215 ms	0.17	0.00	-	-	-	304.74 KB	0.04
AffineSignaturesOnly	500	64	7.119 ms	0.7983 ms	0.0438 ms	0.14	0.00	-	-	-	305.28 KB	0.04
FingerprintsOnly	500	64	10.492 ms	0.6197 ms	0.0340 ms	0.21	0.00	-	-	-	429.7 KB	0.05

ExactPairwise	500	128	49.215 ms	1.2682 ms	0.0695 ms	1.00	0.00	-	-	-	8152.42 KB	1.00
SketchThenVerify	500	128	12.579 ms	6.6481 ms	0.3644 ms	0.26	0.01	218.7500	218.7500	218.7500	4538.22 KB	0.56
SignaturesOnly	500	128	10.279 ms	0.9553 ms	0.0524 ms	0.21	0.00	-	-	-	429.73 KB	0.05
AffineSignaturesOnly	500	128	7.066 ms	0.4505 ms	0.0247 ms	0.14	0.00	-	-	-	430.78 KB	0.05
FingerprintsOnly	500	128	10.306 ms	0.7697 ms	0.0422 ms	0.21	0.00	-	-	-	429.7 KB	0.05

ExactPairwise	2000	64	838.935 ms	266.1737 ms	14.5899 ms	1.00	0.02	1000.0000	-	-	126360.09 KB	1.000
SketchThenVerify	2000	64	40.741 ms	7.3691 ms	0.4039 ms	0.05	0.00	538.4615	538.4615	461.5385	10165.44 KB	0.080
SignaturesOnly	2000	64	34.134 ms	8.5843 ms	0.4705 ms	0.04	0.00	-	-	-	1218.84 KB	0.010
AffineSignaturesOnly	2000	64	27.425 ms	2.3196 ms	0.1271 ms	0.03	0.00	-	-	-	1219.36 KB	0.010
FingerprintsOnly	2000	64	41.053 ms	4.8087 ms	0.2636 ms	0.05	0.00	-	-	-	1718.81 KB	0.014

ExactPairwise	2000	128	845.350 ms	155.6310 ms	8.5307 ms	1.00	0.01	1000.0000	-	-	126359.44 KB	1.00
SketchThenVerify	2000	128	53.639 ms	11.8219 ms	0.6480 ms	0.06	0.00	900.0000	900.0000	800.0000	18495.21 KB	0.15
SignaturesOnly	2000	128	40.191 ms	7.3683 ms	0.4039 ms	0.05	0.00	-	-	-	1718.85 KB	0.01
AffineSignaturesOnly	2000	128	28.039 ms	1.3981 ms	0.0766 ms	0.03	0.00	-	-	-	1719.86 KB	0.01
FingerprintsOnly	2000	128	41.573 ms	1.5665 ms	0.0859 ms	0.05	0.00	-	-	-	1718.81 KB	0.01

Lodestar.Text.Benchmarks.SplitterIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	SampleCount	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar_KFold	10000	70.045 μs	27.0904 μs	1.4849 μs	1.00	0.03	3.2959	1.2207	-	273.94 KB	1.00
Lodestar_StratifiedKFold	10000	145.485 μs	58.6136 μs	3.2128 μs	2.08	0.06	3.6621	1.7090	-	313.53 KB	1.14
Lodestar_TrainTest	10000	4.717 μs	0.3791 μs	0.0208 μs	0.07	0.00	0.4730	0.0458	-	39.14 KB	0.14
MlNet_CrossValidationSplit	10000	244.030 μs	162.2494 μs	8.8934 μs	3.48	0.13	-	-	-	40.55 KB	0.15
MlNet_CrossValidationSplit_Read	10000	3,719.771 μs	97.9481 μs	5.3689 μs	53.12	0.99	-	-	-	148.65 KB	0.54
MlNet_TrainTestSplit	10000	62.482 μs	45.0013 μs	2.4667 μs	0.89	0.03	-	-	-	12.55 KB	0.05
MlNet_TrainTestSplit_Read	10000	672.966 μs	62.0913 μs	3.4034 μs	9.61	0.18	-	-	-	34.17 KB	0.12

Lodestar_KFold	100000	1,035.896 μs	553.4838 μs	30.3383 μs	1.00	0.04	263.6719	259.7656	259.7656	2736.79 KB	1.000
Lodestar_StratifiedKFold	100000	1,656.689 μs	562.4641 μs	30.8306 μs	1.60	0.05	541.0156	537.1094	537.1094	3129.79 KB	1.144
Lodestar_TrainTest	100000	91.887 μs	96.9577 μs	5.3146 μs	0.09	0.00	41.0156	41.0156	41.0156	391.04 KB	0.143
MlNet_CrossValidationSplit	100000	244.842 μs	74.8901 μs	4.1050 μs	0.24	0.01	-	-	-	40.55 KB	0.015
MlNet_CrossValidationSplit_Read	100000	29,561.106 μs	558.2106 μs	30.5974 μs	28.55	0.72	-	-	-	148.71 KB	0.054
MlNet_TrainTestSplit	100000	60.003 μs	6.9828 μs	0.3827 μs	0.06	0.00	-	-	-	12.55 KB	0.005
MlNet_TrainTestSplit_Read	100000	5,160.449 μs	1,392.6108 μs	76.3337 μs	4.98	0.14	-	-	-	34.17 KB	0.012

Lodestar.Text.Benchmarks.StopWordBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Count	200	4.394 ms	2.7165 ms	0.1489 ms	1.00	0.04	109.3750	109.3750	109.3750	1366.79 KB	1.00
CountWithStopWords	200	3.673 ms	0.4303 ms	0.0236 ms	0.84	0.02	7.8125	-	-	725.83 KB	0.53
Hashing	200	4.365 ms	0.7889 ms	0.0432 ms	0.99	0.03	70.3125	70.3125	70.3125	1051.15 KB	0.77
HashingWithStopWords	200	3.885 ms	0.2136 ms	0.0117 ms	0.88	0.03	-	-	-	591.84 KB	0.43

Count	1000	14.012 ms	11.0331 ms	0.6048 ms	1.00	0.05	828.1250	828.1250	828.1250	5880.87 KB	1.00
CountWithStopWords	1000	11.439 ms	0.8453 ms	0.0463 ms	0.82	0.03	375.0000	375.0000	375.0000	3155.02 KB	0.54
Hashing	1000	13.751 ms	0.4813 ms	0.0264 ms	0.98	0.04	531.2500	515.6250	515.6250	4793.15 KB	0.82
HashingWithStopWords	1000	12.667 ms	5.1942 ms	0.2847 ms	0.91	0.04	265.6250	265.6250	265.6250	2633.78 KB	0.45

Lodestar.Text.Benchmarks.TextRankBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Words	Mean	Error	StdDev	Gen0	Gen1	Allocated
Extract	2000	9.090 ms	0.8352 ms	0.0458 ms	31.2500	15.6250	3.13 MB
Extract	8000	221.069 ms	3.5191 ms	0.1929 ms	333.3333	-	35.34 MB

Lodestar.Text.Benchmarks.TokenizerIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Model	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Allocated	Alloc Ratio
Lodestar	WordPiece	15.00 ms	0.487 ms	0.027 ms	1.00	0.00	93.7500	8.71 MB	1.00
MlTokenizers	WordPiece	40.20 ms	0.479 ms	0.026 ms	2.68	0.00	-	3.55 MB	0.41

Lodestar	SentencePiece	36.93 ms	5.627 ms	0.308 ms	1.00	0.01	-	5.44 MB	1.00
MlTokenizers	SentencePiece	44.14 ms	1.366 ms	0.075 ms	1.20	0.01	-	3.09 MB	0.57

Lodestar	ByteLevelBpe	72.17 ms	1.767 ms	0.097 ms	1.00	0.00	285.7143	28.47 MB	1.00
MlTokenizers	ByteLevelBpe	216.84 ms	99.439 ms	5.451 ms	3.00	0.07	666.6667	59.08 MB	2.08

Lodestar.Text.Benchmarks.TopKAccuracyBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Classes	Mean	Error	StdDev	Allocated
TopTwo	10	11.30 ms	4.772 ms	0.262 ms	-
TopTwo	100	80.34 ms	3.546 ms	0.194 ms	-

Lodestar.Text.Benchmarks.VectorMathBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Dim	Mean	Error	StdDev	Ratio	RatioSD	Allocated	Alloc Ratio
Dot	384	36.69 ns	15.240 ns	0.835 ns	1.00	0.03	-	NA
L2Norm	384	30.75 ns	5.910 ns	0.324 ns	0.84	0.02	-	NA

Dot	768	69.73 ns	2.788 ns	0.153 ns	1.00	0.00	-	NA
L2Norm	768	56.70 ns	1.542 ns	0.085 ns	0.81	0.00	-	NA

Dot	1024	91.65 ns	1.403 ns	0.077 ns	1.00	0.00	-	NA
L2Norm	1024	74.74 ns	6.103 ns	0.335 ns	0.82	0.00	-	NA

Lodestar.Text.Benchmarks.VectorizerBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Count	200	2.135 ms	0.5316 ms	0.0291 ms	1.00	0.02	-	-	-	302.94 KB	1.00
Tfidf	200	2.152 ms	0.1734 ms	0.0095 ms	1.01	0.01	3.9063	-	-	332.14 KB	1.10
CountBigrams	200	2.527 ms	0.1477 ms	0.0081 ms	1.18	0.01	3.9063	-	-	590.31 KB	1.95
CountCharWordBoundary	200	1.770 ms	0.3391 ms	0.0186 ms	0.83	0.01	382.8125	382.8125	382.8125	1685.7 KB	5.56
Hashing	200	2.112 ms	0.1026 ms	0.0056 ms	0.99	0.01	-	-	-	233.59 KB	0.77

Count	1000	3.949 ms	0.5961 ms	0.0327 ms	1.00	0.01	109.3750	109.3750	109.3750	1280.03 KB	1.00
Tfidf	1000	4.090 ms	0.6233 ms	0.0342 ms	1.04	0.01	140.6250	140.6250	140.6250	1424.43 KB	1.11
CountBigrams	1000	5.642 ms	1.0713 ms	0.0587 ms	1.43	0.02	398.4375	398.4375	398.4375	2291.91 KB	1.79
CountCharWordBoundary	1000	11.084 ms	1.3093 ms	0.0718 ms	2.81	0.03	1015.6250	1015.6250	1015.6250	12022.21 KB	9.39
Hashing	1000	3.937 ms	0.1037 ms	0.0057 ms	1.00	0.01	70.3125	70.3125	70.3125	1015.48 KB	0.79

Lodestar.Text.Benchmarks.VectorizerIncumbentBenchmarks-report-github

BenchmarkDotNet v0.14.0, Ubuntu 24.04.5 LTS (Noble Numbat)
INTEL XEON PLATINUM 8573C, 1 CPU, 4 logical and 2 physical cores
.NET SDK 10.0.401
  [Host]   : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  ShortRun : .NET 10.0.12 (10.0.1226.42308), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI

Job=ShortRun  IterationCount=3  LaunchCount=1  
WarmupCount=3

Method	Documents	Mean	Error	StdDev	Ratio	RatioSD	Gen0	Gen1	Gen2	Allocated	Alloc Ratio
Lodestar	200	4.644 ms	0.1692 ms	0.0093 ms	1.00	0.00	148.4375	148.4375	148.4375	2.01 MB	1.00
MlNet	200	41.864 ms	33.3280 ms	1.8268 ms	9.01	0.34	7500.0000	7500.0000	7500.0000	28.28 MB	14.10

Lodestar	1000	16.895 ms	2.6893 ms	0.1474 ms	1.00	0.01	1500.0000	1500.0000	1500.0000	9.17 MB	1.00
MlNet	1000	302.545 ms	370.8883 ms	20.3296 ms	17.91	1.05	71000.0000	71000.0000	71000.0000	324.28 MB	35.35

Against rapidfuzz, in this same run

Both sides on this VM in these minutes, which is what makes the ratio readable where the absolutes are not.

levenshtein
indel
metrics
persistence
stats
splitters
glm
var
ols

compare-glm

Python: {'scipy': '1.18.1', 'statsmodels': '0.15.0', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
glm_negative_binomial_n1000	0.313	2.115	6.76x	0.313	2.114	6.76x
glm_gamma_n1000	0.409	2.577	6.30x	0.409	2.577	6.30x
glm_poisson_exposure_n1000	0.327	2.421	7.41x	0.327	2.421	7.41x
mnlogit_n1000	0.687	9.495	13.82x	0.687	9.494	13.82x
glm_negative_binomial_n10000	2.768	8.307	3.00x	2.773	8.306	3.00x
glm_gamma_n10000	3.440	9.912	2.88x	3.444	9.912	2.88x
glm_poisson_exposure_n10000	3.346	9.749	2.91x	3.351	9.748	2.91x
mnlogit_n10000	6.745	49.893	7.40x	6.748	49.891	7.39x
glm_negative_binomial_n100000	29.578	95.243	3.22x	29.773	379.937	12.76x
glm_gamma_n100000	36.433	108.263	2.97x	36.561	432.761	11.84x
glm_poisson_exposure_n100000	42.564	103.502	2.43x	42.780	413.082	9.66x
mnlogit_n100000	67.951	557.270	8.20x	67.992	1226.156	18.03x

ratio > 1 means Lodestar is faster. cpu is the honest one on both sides: elapsed time hides .NET's background GC threads, and numpy's BLAS threads too -- statsmodels at 100 000 rows spends ~10x more cpu than wall.

compare-indel

Python: rapidfuzz 3.14.6 (py 3.12.14)
C#:     Lodestar.Text on .NET 10.0.12 (mode Utf16Unit)

alphabet	length	Python ns/pair	C# ns/pair	speedup (py/C#)
latin	8	91.3	13.8	6.59x C# faster
latin	32	132.1	73.0	1.81x C# faster
latin	128	379.8	312.7	1.21x C# faster
latin	512	3357.5	3910.5	1.16x Py faster
cjk	8	100.6	14.0	7.17x C# faster
cjk	32	246.9	202.1	1.22x C# faster
cjk	128	1665.6	1093.7	1.52x C# faster
cjk	512	14665.4	6909.8	2.12x C# faster

Note: Indel is len(a)+len(b)-2*LCS on both sides, so this compares the subsequence kernels. Lodestar's is Hyyro's bit-parallel LLCS above a pattern of 8 and a rolling-row dynamic program below it (#273).

compare-levenshtein

Python: rapidfuzz 3.14.6 (py 3.12.14)
C#:     Lodestar.Text on .NET 10.0.12 (mode Utf16Unit)

alphabet	length	Python ns/pair	C# ns/pair	speedup (py/C#)
latin	8	112.1	12.8	8.75x C# faster
latin	32	191.7	129.9	1.48x C# faster
latin	128	1307.6	706.8	1.85x C# faster
latin	512	10403.9	9494.2	1.10x C# faster
cjk	8	112.9	12.7	8.90x C# faster
cjk	32	300.9	246.5	1.22x C# faster
cjk	128	2481.1	1796.3	1.38x C# faster
cjk	512	21929.1	15659.5	1.40x C# faster

Note: Python times the realistic per-call loop; rapidfuzz's C core uses the bit-parallel Myers algorithm, so it scales better on long strings.

compare-metrics

Python: {'scikit-learn': '1.9.1', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
confusion_matrix_n1000_k2	0.003	0.646	214.32x	0.003	0.646	214.31x
accuracy_n1000_k2	0.000	0.327	2730.26x	0.000	0.327	2730.02x
precision_recall_f1_macro_n1000_k2	0.003	1.186	409.40x	0.003	1.186	409.37x
classification_report_n1000_k2	0.005	4.399	933.23x	0.005	4.399	933.17x
roc_auc_binary_n1000_k2	0.013	1.318	104.27x	0.013	1.318	104.26x
balanced_accuracy_n1000_k2	0.003	0.697	248.53x	0.003	0.697	248.52x
matthews_n1000_k2	0.003	1.320	470.67x	0.003	1.320	470.66x
cohen_kappa_n1000_k2	0.003	0.728	258.74x	0.003	0.728	258.74x
mse_n1000_k2	0.000	0.178	380.99x	0.000	0.178	381.00x
mae_n1000_k2	0.000	0.178	386.32x	0.000	0.178	386.32x
median_ae_n1000_k2	0.005	0.186	40.16x	0.005	0.186	40.16x
r2_n1000_k2	0.001	0.215	181.54x	0.001	0.215	181.54x
confusion_matrix_n1000_k10	0.003	0.645	201.20x	0.003	0.645	201.17x
accuracy_n1000_k10	0.000	0.327	2755.41x	0.000	0.327	2755.22x
precision_recall_f1_macro_n1000_k10	0.003	1.181	364.01x	0.003	1.181	363.99x
classification_report_n1000_k10	0.008	4.449	583.73x	0.008	4.449	583.69x
roc_auc_ovr_macro_n1000_k10	0.455	6.778	14.91x	0.455	6.777	14.91x
balanced_accuracy_n1000_k10	0.003	0.697	226.62x	0.003	0.697	226.62x
matthews_n1000_k10	0.003	1.339	437.37x	0.003	1.339	437.33x
cohen_kappa_n1000_k10	0.003	0.736	223.55x	0.003	0.736	223.54x
mse_n1000_k10	0.000	0.177	384.20x	0.000	0.177	384.16x
mae_n1000_k10	0.000	0.177	387.30x	0.000	0.177	387.29x
median_ae_n1000_k10	0.005	0.189	40.79x	0.005	0.189	40.79x
r2_n1000_k10	0.001	0.213	181.50x	0.001	0.213	181.48x
confusion_matrix_n100000_k2	0.288	10.058	34.90x	0.288	10.057	34.90x
accuracy_n100000_k2	0.013	3.469	261.41x	0.013	3.469	261.40x
precision_recall_f1_macro_n100000_k2	0.273	11.421	41.82x	0.273	11.421	41.82x
classification_report_n100000_k2	0.275	24.303	88.27x	0.275	24.299	88.26x
roc_auc_binary_n100000_k2	3.503	23.293	6.65x	3.503	23.293	6.65x
balanced_accuracy_n100000_k2	0.273	10.116	37.06x	0.273	10.116	37.06x
matthews_n100000_k2	0.272	20.419	75.20x	0.272	20.417	75.18x
cohen_kappa_n100000_k2	0.273	10.200	37.33x	0.273	10.200	37.33x
mse_n100000_k2	0.044	0.418	9.46x	0.044	0.418	9.46x
mae_n100000_k2	0.044	0.411	9.43x	0.044	0.410	9.43x
median_ae_n100000_k2	0.518	1.529	2.95x	0.526	1.529	2.91x
r2_n100000_k2	0.112	0.647	5.76x	0.112	0.646	5.76x
confusion_matrix_n100000_k10	0.255	10.041	39.31x	0.255	10.037	39.29x
accuracy_n100000_k10	0.013	3.479	261.04x	0.013	3.479	261.02x
precision_recall_f1_macro_n100000_k10	0.254	11.919	46.94x	0.254	11.919	46.95x
classification_report_n100000_k10	0.263	26.269	99.96x	0.263	26.267	99.96x
roc_auc_ovr_macro_n100000_k10	35.549	182.085	5.12x	35.547	182.076	5.12x
balanced_accuracy_n100000_k10	0.254	10.087	39.69x	0.254	10.087	39.68x
matthews_n100000_k10	0.256	20.883	81.53x	0.256	20.881	81.53x
cohen_kappa_n100000_k10	0.279	10.108	36.28x	0.279	10.107	36.29x
mse_n100000_k10	0.046	0.406	8.90x	0.046	0.406	8.90x
mae_n100000_k10	0.045	0.398	8.84x	0.045	0.398	8.84x
median_ae_n100000_k10	0.556	1.502	2.70x	0.592	1.502	2.54x
r2_n100000_k10	0.116	0.627	5.38x	0.116	0.627	5.38x
confusion_matrix_n1000000_k2	2.722	95.475	35.07x	2.722	95.445	35.06x
accuracy_n1000000_k2	0.226	31.604	140.09x	0.226	31.602	140.09x
precision_recall_f1_macro_n1000000_k2	2.711	102.816	37.93x	2.711	102.792	37.92x
classification_report_n1000000_k2	2.728	199.592	73.17x	2.728	199.577	73.17x
roc_auc_binary_n1000000_k2	58.133	260.570	4.48x	58.130	260.561	4.48x
balanced_accuracy_n1000000_k2	2.697	95.895	35.56x	2.697	95.891	35.56x
matthews_n1000000_k2	2.703	191.153	70.71x	2.703	191.143	70.71x
cohen_kappa_n1000000_k2	2.725	96.221	35.31x	2.725	96.216	35.31x
mse_n1000000_k2	0.494	2.386	4.83x	0.494	2.386	4.83x
mae_n1000000_k2	0.488	2.377	4.87x	0.488	2.377	4.87x
median_ae_n1000000_k2	5.515	12.478	2.26x	5.569	12.477	2.24x
r2_n1000000_k2	1.165	5.006	4.30x	1.165	5.006	4.30x
confusion_matrix_n1000000_k10	2.526	95.748	37.90x	2.526	95.717	37.89x
accuracy_n1000000_k10	0.214	31.640	148.12x	0.214	31.639	148.12x
precision_recall_f1_macro_n1000000_k10	2.528	109.109	43.15x	2.528	109.086	43.15x
classification_report_n1000000_k10	2.561	218.977	85.51x	2.560	218.950	85.51x
balanced_accuracy_n1000000_k10	2.525	96.645	38.27x	2.525	96.637	38.28x
matthews_n1000000_k10	2.530	198.810	78.58x	2.530	198.801	78.58x
cohen_kappa_n1000000_k10	2.524	96.402	38.20x	2.524	96.394	38.20x
mse_n1000000_k10	0.512	2.398	4.69x	0.512	2.398	4.69x
mae_n1000000_k10	0.505	2.393	4.74x	0.505	2.393	4.74x
median_ae_n1000000_k10	4.942	12.513	2.53x	4.968	12.512	2.52x
r2_n1000000_k10	1.184	5.024	4.24x	1.184	5.024	4.24x

ratio > 1 means Lodestar is faster. cpu is the merge gate for this branch (docs/guides/performance.md): every operation, every size, must be >= 1x.

compare-ols

Python: {'scipy': '1.18.1', 'statsmodels': '0.15.0', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
ols_summary_n1000	0.043	1.618	37.70x	0.043	1.618	37.69x
ols_hac_n1000	0.155	2.070	13.35x	0.155	2.070	13.36x
ols_cluster_n1000	0.059	2.112	36.07x	0.059	2.111	36.06x
ols_summary_n10000	0.419	7.791	18.60x	0.419	7.791	18.60x
ols_hac_n10000	1.558	8.787	5.64x	1.560	8.786	5.63x
ols_cluster_n10000	0.562	9.128	16.23x	0.563	9.127	16.20x
ols_summary_n100000	4.103	89.042	21.70x	4.112	355.410	86.43x
ols_hac_n100000	15.908	100.229	6.30x	16.076	400.243	24.90x
ols_cluster_n100000	6.068	102.016	16.81x	6.168	407.660	66.09x

compare-persistence

Python: {'tokenizers': '0.23.2', 'sentencepiece': '0.2.2', 'scikit-learn': '1.9.1', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu	C# bytes	Py bytes
vocab_txt	2.834	8.266	2.92x	2.955	8.265	2.80x	228,891	228,891
tokenizer_json_wordpiece	7.002	13.541	1.93x	7.177	13.540	1.89x	706,526	706,526
tokenizer_json_unigram	9.346	31.931	3.42x	10.174	31.931	3.14x	1,990,038	1,990,038
spiece_model	2.986	25.432	8.52x	3.154	25.431	8.06x	533,084	533,084
tfidf_save	1.543	2.130	1.38x	1.555	2.130	1.37x	581,787	591,922
tfidf_load	3.201	3.492	1.09x	3.333	3.492	1.05x	581,787	591,922
embedding_index_save	4.633	1.957	0.42x	4.802	1.957	0.41x	20,589,007	15,360,128
embedding_index_save_file	81.060	70.503	0.87x	8.254	4.902	0.59x	20,589,007	15,360,128
embedding_index_load	6.101	1.224	0.20x	6.472	1.224	0.19x	20,589,007	15,360,128
embedding_index_load_file	7.506	1.030	0.14x	7.893	1.024	0.13x	20,589,007	15,360,128
embedding_index_load_memory	4.722	1.224	0.26x	5.018	1.224	0.24x	20,589,007	15,360,128
embedding_index_ingest_npy	1.951	1.223	0.63x	2.105	1.223	0.58x	15,360,128	15,360,128
embedding_index_view_floor	0.000	0.000	82.11x	0.000	0.000	82.10x	20,589,007	15,360,128
embedding_index_save_gzip	365.920	521.716	1.43x	365.896	521.687	1.43x	15,250,490	14,022,374
embedding_index_load_gzip	67.857	79.488	1.17x	68.230	79.484	1.16x	15,250,490	14,022,374

ratio > 1 means Lodestar is faster. cpu is the honest one: elapsed time hides work .NET does on background GC threads; CPython is single-threaded. bytes is what the row wrote or read; a results file from before #378 carries none, and the two columns then disappear rather than read zero.

compare-splitters

Python: {'scikit-learn': '1.9.1', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
kfold_n10000	0.081	0.086	1.06x	0.081	0.086	1.06x
stratified_n10000	0.134	0.644	4.79x	0.134	0.644	4.79x
traintest_n10000	0.004	0.154	34.41x	0.004	0.154	34.41x
kfold_n100000	1.064	1.838	1.73x	1.128	1.838	1.63x
stratified_n100000	1.482	6.521	4.40x	1.559	6.521	4.18x
traintest_n100000	0.128	0.386	3.02x	0.129	0.386	3.00x
kfold_n1000000	10.734	13.525	1.26x	11.838	13.520	1.14x
stratified_n1000000	16.057	60.884	3.79x	16.795	60.880	3.62x
traintest_n1000000	0.766	2.445	3.19x	0.844	2.445	2.90x

compare-stats

Python: {'scipy': '1.18.1', 'statsmodels': '0.15.0', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
welch_t_n1000	0.002	0.478	196.48x	0.002	0.478	196.46x
mann_whitney_n1000	0.043	0.506	11.86x	0.043	0.506	11.86x
chi_square_n1000	0.000	0.188	1277.82x	0.000	0.188	1277.80x
welch_t_n10000	0.023	0.523	23.01x	0.023	0.523	23.01x
mann_whitney_n10000	0.494	2.108	4.27x	0.494	2.108	4.27x
chi_square_n10000	0.001	0.195	206.95x	0.001	0.195	206.96x
welch_t_n100000	0.225	1.044	4.63x	0.225	1.044	4.63x
mann_whitney_n100000	5.653	21.800	3.86x	5.652	21.799	3.86x
chi_square_n100000	0.010	0.204	21.20x	0.010	0.204	21.20x

compare-var

Python: {'scipy': '1.18.1', 'statsmodels': '0.15.0', 'numpy': '2.5.3'} (py 3.12.14)
C#:     Lodestar on .NET 10.0.12

operation	C# ms	Py ms	wall	C# cpu	Py cpu	cpu
var_n1000	0.049	1.808	36.77x	0.049	1.808	36.77x
var_n10000	0.563	10.987	19.51x	0.565	10.987	19.44x
var_n100000	5.465	102.518	18.76x	5.572	102.514	18.40x

Selected, and not reached tonight

The run stops starting classes when the budget left cannot hold one, so these were carried to the next run rather than measured badly or killed mid-flight. They are selected again tomorrow whether or not anything else changes.

BitParallelEditDistanceBenchmarks
ChainedProductBenchmarks
CoxBenchmarks
GlmBenchmarks
GlmNegativeBinomialBenchmarks
GlmOffsetBenchmarks
GlmPoissonBenchmarks
GlsBenchmarks
HacClusterBenchmarks
KolmogorovDurbinBenchmarks
KsExactTableBenchmarks
LeastSquaresRoutingBenchmarks
MannWhitneyExactBenchmarks
MultinomialLogitBenchmarks
OlsBenchmarks
QuantileBenchmarks
RobustCovarianceBenchmarks
SerialCorrelationBenchmarks
SurvivalBenchmarks
TiledCosineTopKBenchmarks
TiledMinHashSignaturesBenchmarks
TiledSparseDenseProductBenchmarks
VectorAutoregressionBenchmarks
WeightedLeastSquaresBenchmarks

Ratios that moved

Read against bench/nightly/ratios.csv by tools/nightly_series.py, under the thresholds that script sets. A ratio moves when either side of it does: read its baseline before calling a movement a regression.

New tonight: 38. Each of these had not moved on its previous reading.

class	parameters	method	kind	tonight	against	change	threshold
`PartialFitBenchmarks`	RowCount=100000, BatchCount=10	`SparseFit`	step	0.55	0.06	+817%	30%
`PartialFitBenchmarks`	RowCount=100000, BatchCount=100	`SparseFit`	step	0.52	0.06	+767%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=100	`SparseFit`	step	0.41	0.06	+583%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=10	`SparseFit`	step	0.36	0.06	+500%	30%
`PartialFitBenchmarks`	RowCount=100000, BatchCount=10	`DenseFitOfTheSameData`	step	1.5	0.69	+117%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=100	`DenseFitOfTheSameData`	step	1.44	0.67	+115%	30%
`EncoderIncumbentBenchmarks`	Job=ShortRun, InvocationCount=Default, IterationCount=3, RunStrategy=Default, UnrollFactor=16, WarmupCount=3, RowCount=1000	`MlNet_OneHotEncoding_Fit`	step	12.73	5.93	+115%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=10	`DenseFitOfTheSameData`	step	1.25	0.67	+87%	30%
`PartialFitBenchmarks`	RowCount=100000, BatchCount=100	`DenseFitOfTheSameData`	step	1.23	0.68	+81%	30%
`StatsBenchmarks`	SampleSize=10000	`LodestarMannWhitney`	step	19.491	10.8775	+79%	30%
`SplitterIncumbentBenchmarks`	SampleCount=10000	`MlNet_TrainTestSplit`	step	0.89	0.54	+65%	30%
`OsaBenchmarks`	Length=8	`Distance_CodePoint`	step	4.94	3.01	+64%	30%
`PartialFitBenchmarks`	RowCount=100000, BatchCount=10	`BatchedFit`	step	1.33	0.85	+56%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=10	`BatchedFit`	step	1.27	0.83	+53%	30%
`SplitterIncumbentBenchmarks`	SampleCount=100000	`MlNet_TrainTestSplit`	step	0.06	0.04	+50%	30%
`StatsBenchmarks`	SampleSize=10000	`AccordMannWhitney`	step	495.542	330.894	+50%	30%
`LcsGateBenchmarks`	Band=20	`Kernel`	step	0.16	0.11	+45%	30%
`KMeansFitIncumbentBenchmarks`	Shape=10000x16x16	`MetaNumerics_Fit`	step	8.47	15.345	-45%	30%
`PartialFitBenchmarks`	RowCount=10000, BatchCount=100	`BatchedFit`	step	1.92	1.35	+42%	30%
`KMeansFitIncumbentBenchmarks`	Shape=50000x8x32	`MetaNumerics_Fit`	step	14.05	24.035	-42%	30%
`LcsGateBenchmarks`	Band=20	`Kernel_Cjk`	step	0.24	0.17	+41%	30%
`SplitterIncumbentBenchmarks`	SampleCount=10000	`Lodestar_TrainTest`	step	0.07	0.05	+40%	30%
`StatsBenchmarks`	SampleSize=10000	`AccordChiSquare`	step	0.007	0.005	+40%	30%
`StatsBenchmarks`	SampleSize=100	`AccordMannWhitney`	step	30.85	22.12	+39%	30%
`DbscanDimensionBenchmarks`	Shape=5000x16x8	`NumFlat_Fit`	step	8.49	6.22	+36%	30%
`MetricsIncumbentBenchmarks`	Samples=1000000, Request=Bundle	`MlNet`	step	1.48	2.28	-35%	30%
`StatsBenchmarks`	SampleSize=10000	`LodestarChiSquare`	step	0.004	0.003	+33%	30%
`BkTreeBenchmarks`	Radius=2, Shape=clustered	`TreeWithinDistance`	step	1.67	1.26	+33%	30%
`PartialFitBenchmarks`	RowCount=100000, BatchCount=100	`BatchedFit`	step	1.43	1.08	+32%	30%
`KMeansFitIncumbentBenchmarks`	Shape=10000x2x8	`NumFlat_Fit`	step	5.23	7.635	-31%	30%
`DistributionTailBenchmarks`	—	`NormalQuantile`	step	5.11	3.89	+31%	30%
`LevenshteinCodePointBenchmarks`	Length=512, Distinct=512	`Distance_Utf16`	step	0.11	0.16	-31%	30%
`LevenshteinIncumbentBenchmarks`	Length=64	`Quickenshtein`	step	5.35	4.08	+31%	31%
`AgglomerativeIncumbentBenchmarks`	Rows=500, Method=single	`Aglomera_GetClustering`	step	138.55	198.36	-30%	30%
`BkTreeBenchmarks`	Radius=4, Shape=uniform	`TreeWithinDistance`	step	1.77	1.36	+30%	30%
`KMeansLloydIncumbentBenchmarks`	Shape=10000x2x8	`NumFlat_Lloyd`	step	2.97	4.245	-30%	30%
`LevenshteinIncumbentBenchmarks`	Length=512	`F23_StringSimilarity`	drift	69.7	50.37	+26%	20%
`LevenshteinCodePointBenchmarks`	Length=512, Distinct=32	`Distance_Utf16`	drift	4.12	3.41	+21%	20%

Still away from their median: 38. These moved on an earlier run, and their median has not caught up yet.

class	parameters	method	kind	tonight	against	change	threshold
`MetricsIncumbentBenchmarks`	Samples=100000, Request=AccuracyAlone	`MlNet`	drift	2249.26	147.17	+2082%	20%
`FuzzIncumbentBenchmarks`	Operation=PartialRatio	`FuzzySharp`	drift	14.29	0.85	+1459%	20%
`TokenizerIncumbentBenchmarks`	Model=SentencePiece	`MlTokenizers`	drift	1.2	0.15	+700%	20%
`IndelBenchmarks`	Length=512	`Distance_CodePoint`	drift	1.07	43.93	-98%	20%
`IndelBenchmarks`	Length=128	`Distance_CodePoint`	drift	1.07	22.68	-95%	20%
`FuzzIncumbentBenchmarks`	Operation=WRatio	`FuzzySharp`	drift	4.27	2.19	+95%	20%
`FuzzBenchmarks`	—	`PartialRatio`	drift	7.24	117.71	-94%	20%
`IndelBenchmarks`	Length=32	`Distance_CodePoint`	drift	1.02	16.61	-94%	20%
`IndelBenchmarks`	Length=24	`Distance_CodePoint`	drift	1.19	12.86	-91%	20%
`IndelBenchmarks`	Length=20	`Distance_CodePoint`	drift	1.25	5.22	-78%	20%
`IndelBenchmarks`	Length=12	`Distance_CodePoint`	drift	1.27	5.26	-76%	20%
`IndelBenchmarks`	Length=16	`Distance_CodePoint`	drift	1.38	5.16	-74%	20%
`IndelBenchmarks`	Length=8	`Distance_CodePoint`	drift	1.32	4.82	-73%	20%
`TokenizerIncumbentBenchmarks`	Model=WordPiece	`MlTokenizers`	step	2.68	1.56	+72%	30%
`MetricsIncumbentBenchmarks`	Samples=1000000, Request=AccuracyAlone	`MlNet`	step	832.69	2082.84	-60%	38%
`VectorizerIncumbentBenchmarks`	Documents=1000	`MlNet`	drift	17.91	12.89	+56%	20%
`BpeBenchmarks`	—	`Bpe`	drift	3.02	1.71	+56%	20%
`FuzzBenchmarks`	—	`WRatio`	drift	13.21	23.78	-54%	20%
`LcsGateBenchmarks`	Band=96	`Kernel`	drift	0.03	0.06	-50%	20%
`VectorizerIncumbentBenchmarks`	Documents=200	`MlNet`	drift	9.01	7.09	+48%	20%
`DecompositionBenchmarks`	—	`MlNet_ProjectToPrincipalComponents_Rank20`	drift	0.93	0.84	+44%	20%
`MyersGateBenchmarks`	Band=96	`Kernel`	drift	0.06	0.1	-40%	20%
`BatchEmbeddingBenchmarks`	CorpusSize=32	`EmbedBatchBucketed`	drift	0.31	0.54	-39%	20%
`MetricsIncumbentBenchmarks`	Samples=100000, Request=Bundle	`MlNet`	drift	4.05	4.17	+33%	20%
`LevenshteinCodePointBenchmarks`	Length=128, Distinct=512	`Distance_Utf16`	drift	2.58	2.11	+33%	20%
`BatchEmbeddingBenchmarks`	CorpusSize=8	`EmbedBatchBucketed`	step	0.35	0.52	-33%	30%
`BatchEmbeddingBenchmarks`	CorpusSize=8	`EmbedBatch`	step	0.35	0.51	-31%	30%
`BatchEmbeddingBenchmarks`	CorpusSize=128	`EmbedBatch`	step	0.33	0.48	-31%	30%
`BatchEmbeddingBenchmarks`	CorpusSize=128	`EmbedBatchBucketed`	step	0.29	0.42	-31%	30%
`LevenshteinCodePointBenchmarks`	Length=128, Distinct=32	`Distance_Utf16`	drift	2.71	2.14	+31%	20%
`BatchEmbeddingBenchmarks`	CorpusSize=32	`EmbedBatch`	step	0.34	0.49	-31%	30%
`LevenshteinIncumbentBenchmarks`	Length=512	`Fastenshtein`	drift	40.88	29.94	+30%	20%
`LcsGateBenchmarks`	Band=18	`Kernel_Cjk`	drift	0.28	0.28	-29%	20%
`MyersGateBenchmarks`	Band=64	`Kernel_Cjk`	drift	0.08	0.07	+29%	20%
`FuzzBenchmarks`	—	`TokenSortRatio`	drift	9.37	10.79	-25%	20%
`FuzzBenchmarks`	—	`TokenSetRatio`	drift	10.13	11.76	-24%	20%
`LcsGateBenchmarks`	Band=18	`Kernel`	drift	0.18	0.17	-24%	20%
`DecompositionBenchmarks`	—	`Nmf_Rank20`	drift	5.58	7.17	-22%	20%