matrixStats.benchmarks

---

indexByRow() benchmarks

This report benchmark the performance of indexByRow() against alternative methods:

• indexByRow_R1() based in matrix(..., byrow = TRUE)
• indexByRow_R2() is a modified version of indexByRow_R1()

where indexByRow_R1() and indexByRow_R2() are defined as in the Appendix.

Data

> data <- rmatrices(mode = "index")

where rmatrices() is defined in the Appendix.

Results

10x10 matrix

> X <- data[["10x10"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:100] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:50] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.001282	0.0016435	0.0018852	0.0017985	0.001945	0.009008
3	indexByRow_R2	0.006167	0.0068870	0.0071962	0.0071035	0.007310	0.013159
2	indexByRow_R1	0.006154	0.0069255	0.0074135	0.0071710	0.007465	0.028892

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	indexByRow_R2	4.810452	4.190447	3.817218	3.949680	3.758355	1.460813
2	indexByRow_R1	4.800312	4.213873	3.932485	3.987212	3.838046	3.207371

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.002608	0.0028595	0.0030970	0.0030175	0.0031545	0.010545
2	indexByRow_R1	0.006912	0.0074970	0.0080785	0.0077735	0.0081135	0.025745
3	indexByRow_R2	0.008396	0.0090915	0.0096742	0.0093155	0.0096055	0.032637

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	2.650307	2.621787	2.608511	2.576139	2.572040	2.441441
3	indexByRow_R2	3.219325	3.179402	3.123750	3.087158	3.045015	3.095021

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.002114	0.0024910	0.0028393	0.0026600	0.0028180	0.018345
3	indexByRow_R2	0.005747	0.0066405	0.0069985	0.0068935	0.0071700	0.016073
2	indexByRow_R1	0.006692	0.0075480	0.0081590	0.0078975	0.0082915	0.028833

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	indexByRow_R2	2.718543	2.665797	2.464875	2.591541	2.544358	0.8761515
2	indexByRow_R1	3.165563	3.030108	2.873589	2.968985	2.942335	1.5717089

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x10+odd data. Outliers are displayed as crosses. Times are in milliseconds.

100x100 matrix

> X <- data[["100x100"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:10000] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:5000] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.015252	0.0161600	0.0177820	0.0172175	0.0179435	0.065262
3	indexByRow_R2	0.073665	0.0769315	0.0826033	0.0819060	0.0866595	0.153746
2	indexByRow_R1	0.074657	0.0783730	0.0837017	0.0844410	0.0866765	0.108103

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	indexByRow_R2	4.829858	4.760613	4.645338	4.757137	4.829576	2.355827
2	indexByRow_R1	4.894899	4.849814	4.707109	4.904371	4.830524	1.656446

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	0.073030	0.0864855	0.0963237	0.0928365	0.104936	0.156011
1	indexByRow	0.081498	0.0924100	0.1052432	0.1017470	0.119113	0.155555
3	indexByRow_R2	0.299841	0.3473715	0.3912514	0.3782655	0.427674	0.531842

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	indexByRow	1.115952	1.068503	1.092599	1.095981	1.135101	0.9970771
3	indexByRow_R2	4.105724	4.016529	4.061838	4.074534	4.075570	3.4090032

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.055356	0.0599685	0.0660457	0.0620160	0.0713825	0.090670
2	indexByRow_R1	0.087818	0.0946170	0.1042238	0.0997865	0.1118695	0.135953
3	indexByRow_R2	0.177048	0.1883325	0.2090049	0.1999050	0.2248520	0.280346

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	1.586422	1.577778	1.578057	1.609044	1.567184	1.499426
3	indexByRow_R2	3.198352	3.140524	3.164550	3.223442	3.149960	3.091938

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x100+odd data. Outliers are displayed as crosses. Times are in milliseconds.

1000x10 matrix

> X <- data[["1000x10"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:10000] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:5000] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.014092	0.0159260	0.0167545	0.0167715	0.0172140	0.024884
3	indexByRow_R2	0.072366	0.0781810	0.0828228	0.0843585	0.0854785	0.098342
2	indexByRow_R1	0.072171	0.0780595	0.0832589	0.0845390	0.0856575	0.115141

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	indexByRow_R2	5.135254	4.909017	4.943302	5.029872	4.965638	3.952017
2	indexByRow_R1	5.121416	4.901388	4.969332	5.040634	4.976037	4.627110

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	0.074761	0.0810275	0.0915757	0.0891510	0.0978105	0.127025
1	indexByRow	0.083101	0.0937025	0.1042172	0.1002460	0.1129245	0.143002
3	indexByRow_R2	0.316553	0.3533775	0.3849551	0.3771155	0.4034030	0.546913

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	indexByRow	1.111555	1.156428	1.138045	1.124452	1.154523	1.125778
3	indexByRow_R2	4.234200	4.361205	4.203683	4.230076	4.124332	4.305554

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.058981	0.0608665	0.0665945	0.0641555	0.0715370	0.087223
2	indexByRow_R1	0.091585	0.0955840	0.1065472	0.1032885	0.1176660	0.176565
3	indexByRow_R2	0.187444	0.1895135	0.2122267	0.2009670	0.2389775	0.279790

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	1.552788	1.570388	1.599940	1.609971	1.644827	2.024294
3	indexByRow_R2	3.178040	3.113593	3.186852	3.132498	3.340614	3.207755

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x10+odd data. Outliers are displayed as crosses. Times are in milliseconds.

10x1000 matrix

> X <- data[["10x1000"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:10000] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:5000] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.014967	0.0160025	0.0167037	0.0165680	0.0169950	0.032126
3	indexByRow_R2	0.076358	0.0782465	0.0830444	0.0822775	0.0857780	0.103863
2	indexByRow_R1	0.076306	0.0795235	0.0835379	0.0830110	0.0857735	0.112211

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	indexByRow_R2	5.101757	4.889642	4.971619	4.966049	5.047249	3.232989
2	indexByRow_R1	5.098283	4.969442	5.001164	5.010321	5.046984	3.492841

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	0.072936	0.0834525	0.0937125	0.0920540	0.1014270	0.130469
1	indexByRow	0.084448	0.0916775	0.1042213	0.1025965	0.1112685	0.142197
3	indexByRow_R2	0.307719	0.3373145	0.3758221	0.3685620	0.4040380	0.494355

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	indexByRow	1.157837	1.098559	1.112139	1.114525	1.097030	1.089891
3	indexByRow_R2	4.219028	4.041994	4.010373	4.003759	3.983535	3.789061

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.057493	0.0612405	0.0653401	0.0621600	0.0701970	0.088982
2	indexByRow_R1	0.087120	0.0938180	0.1042518	0.0985685	0.1142000	0.183106
3	indexByRow_R2	0.174264	0.1841790	0.1988696	0.1904060	0.2120615	0.250581

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	1.515315	1.531960	1.595525	1.585722	1.626850	2.057787
3	indexByRow_R2	3.031047	3.007471	3.043607	3.063160	3.020948	2.816086

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 10x1000+odd data. Outliers are displayed as crosses. Times are in milliseconds.

100x1000 matrix

> X <- data[["100x1000"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:100000] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:50000] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.099124	0.1175865	0.1505342	0.1324315	0.2031230	0.220248
3	indexByRow_R2	0.525359	0.6359585	0.8410909	0.6781020	0.8273405	12.569268
2	indexByRow_R1	0.524381	0.6389070	0.7295728	0.7045310	0.8274250	1.127695

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	indexByRow_R2	5.300018	5.408431	5.587372	5.120398	4.073101	57.068704
2	indexByRow_R1	5.290152	5.433506	4.846557	5.319965	4.073517	5.120115

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	0.676463	0.684153	0.9069105	0.6973605	0.9944955	7.065740
1	indexByRow	0.779384	0.782105	0.8616909	0.8484975	0.9002355	1.303930
3	indexByRow_R2	3.012132	3.031932	3.4665338	3.2426560	3.9203275	5.297835

	expr	min	lq	mean	median	uq	max
2	indexByRow_R1	1.000000	1.000000	1.0000000	1.000000	1.0000000	1.0000000
1	indexByRow	1.152146	1.143173	0.9501388	1.216727	0.9052183	0.1845426
3	indexByRow_R2	4.452767	4.431658	3.8223547	4.649899	3.9420264	0.7497920

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.458584	0.4633480	0.5287863	0.529181	0.579864	0.710381
2	indexByRow_R1	0.691888	0.7261385	0.8803531	0.865033	1.029082	1.513058
3	indexByRow_R2	1.469839	1.4964135	1.8339744	1.828790	1.928001	8.186444

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	1.508749	1.567156	1.664856	1.634664	1.774695	2.129925
3	indexByRow_R2	3.205169	3.229567	3.468271	3.455887	3.324919	11.524019

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 100x1000+odd data. Outliers are displayed as crosses. Times are in milliseconds.

1000x100 matrix

> X <- data[["1000x100"]]
> dim <- dim(X)
> idxsList <- list(`all-by-NULL` = NULL, all = seq_len(prod(dim)), odd = seq(from = 1, to = prod(dim), 
+     by = 2L))
> str(idxsList)
List of 3
 $ all-by-NULL: NULL
 $ all        : int [1:100000] 1 2 3 4 5 6 7 8 9 10 ...
 $ odd        : num [1:50000] 1 3 5 7 9 11 13 15 17 19 ...

Index set ‘all-by-NULL’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+all-by-NULL data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.108854	0.1236085	0.2695321	0.1354835	0.2066770	11.379597
2	indexByRow_R1	0.510801	0.6014975	0.6972782	0.6633375	0.8146630	0.875310
3	indexByRow_R2	0.511998	0.6085370	0.7112154	0.7047580	0.8220385	0.870048

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
2	indexByRow_R1	4.692533	4.86615	2.586995	4.896076	3.941721	0.0769192
3	indexByRow_R2	4.703530	4.92310	2.638704	5.201799	3.977407	0.0764568

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+all-by-NULL data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘all’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+all data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.779206	0.797280	0.8907752	0.894966	0.910122	1.359681
2	indexByRow_R1	0.659909	0.714930	0.8888209	0.965549	1.003583	1.629828
3	indexByRow_R2	2.993995	3.138649	3.8375815	3.901712	3.962118	15.189870

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.0000000	1.0000000	1.0000000	1.000000	1.000000	1.000000
2	indexByRow_R1	0.8468993	0.8967113	0.9978061	1.078867	1.102691	1.198684
3	indexByRow_R2	3.8423665	3.9366960	4.3081367	4.359621	4.353392	11.171643

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+all data. Outliers are displayed as crosses. Times are in milliseconds.

Index set ‘odd’

> stats <- microbenchmark(indexByRow = indexByRow(dim, idxs = idxs), indexByRow_R1 = indexByRow_R1(dim, 
+     idxs = idxs), indexByRow_R2 = indexByRow_R2(dim, idxs = idxs), unit = "ms")

Table: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+odd data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	indexByRow	0.459075	0.4646340	0.5462618	0.5710475	0.5814035	0.942403
2	indexByRow_R1	0.703373	0.7178475	0.8790656	0.8398025	1.0195915	2.193599
3	indexByRow_R2	1.489537	1.4968780	1.7872100	1.6315585	1.9249515	8.194279

	expr	min	lq	mean	median	uq	max
1	indexByRow	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	indexByRow_R1	1.532153	1.544974	1.609239	1.470635	1.753673	2.327666
3	indexByRow_R2	3.244648	3.221628	3.271710	2.857133	3.310870	8.695090

Figure: Benchmarking of indexByRow(), indexByRow_R1() and indexByRow_R2() on 1000x100+odd data. Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

R version 4.1.1 Patched (2021-08-10 r80727)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.5 LTS

Matrix products: default
BLAS:   /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRblas.so
LAPACK: /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] microbenchmark_1.4-7   matrixStats_0.60.0     ggplot2_3.3.5         
[4] knitr_1.33             R.devices_2.17.0       R.utils_2.10.1        
[7] R.oo_1.24.0            R.methodsS3_1.8.1-9001 history_0.0.1-9000    

loaded via a namespace (and not attached):
 [1] Biobase_2.52.0          httr_1.4.2              splines_4.1.1          
 [4] bit64_4.0.5             network_1.17.1          assertthat_0.2.1       
 [7] highr_0.9               stats4_4.1.1            blob_1.2.2             
[10] GenomeInfoDbData_1.2.6  robustbase_0.93-8       pillar_1.6.2           
[13] RSQLite_2.2.8           lattice_0.20-44         glue_1.4.2             
[16] digest_0.6.27           XVector_0.32.0          colorspace_2.0-2       
[19] Matrix_1.3-4            XML_3.99-0.7            pkgconfig_2.0.3        
[22] zlibbioc_1.38.0         genefilter_1.74.0       purrr_0.3.4            
[25] ergm_4.1.2              xtable_1.8-4            scales_1.1.1           
[28] tibble_3.1.4            annotate_1.70.0         KEGGREST_1.32.0        
[31] farver_2.1.0            generics_0.1.0          IRanges_2.26.0         
[34] ellipsis_0.3.2          cachem_1.0.6            withr_2.4.2            
[37] BiocGenerics_0.38.0     mime_0.11               survival_3.2-13        
[40] magrittr_2.0.1          crayon_1.4.1            statnet.common_4.5.0   
[43] memoise_2.0.0           laeken_0.5.1            fansi_0.5.0            
[46] R.cache_0.15.0          MASS_7.3-54             R.rsp_0.44.0           
[49] progressr_0.8.0         tools_4.1.1             lifecycle_1.0.0        
[52] S4Vectors_0.30.0        trust_0.1-8             munsell_0.5.0          
[55] tabby_0.0.1-9001        AnnotationDbi_1.54.1    Biostrings_2.60.2      
[58] compiler_4.1.1          GenomeInfoDb_1.28.1     rlang_0.4.11           
[61] grid_4.1.1              RCurl_1.98-1.4          cwhmisc_6.6            
[64] rstudioapi_0.13         rappdirs_0.3.3          startup_0.15.0-9000    
[67] labeling_0.4.2          bitops_1.0-7            base64enc_0.1-3        
[70] boot_1.3-28             gtable_0.3.0            DBI_1.1.1              
[73] markdown_1.1            R6_2.5.1                lpSolveAPI_5.5.2.0-17.7
[76] rle_0.9.2               dplyr_1.0.7             fastmap_1.1.0          
[79] bit_4.0.4               utf8_1.2.2              parallel_4.1.1         
[82] Rcpp_1.0.7              vctrs_0.3.8             png_0.1-7              
[85] DEoptimR_1.0-9          tidyselect_1.1.1        xfun_0.25              
[88] coda_0.19-4            

Total processing time was 19.77 secs.

Reproducibility

To reproduce this report, do:

html <- matrixStats:::benchmark('indexByRow')

---

Copyright Henrik Bengtsson. Last updated on 2021-08-25 22:35:02 (+0200 UTC). Powered by RSP.

Local functions

> indexByRow_R1 <- function(dim, idxs = NULL, ...) {
+     n <- prod(dim)
+     x <- matrix(seq_len(n), nrow = dim[2L], ncol = dim[1L], byrow = TRUE)
+     if (!is.null(idxs)) 
+         x <- x[idxs]
+     as.vector(x)
+ }

> indexByRow_R2 <- function(dim, idxs = NULL, ...) {
+     n <- prod(dim)
+     if (is.null(idxs)) {
+         x <- matrix(seq_len(n), nrow = dim[2L], ncol = dim[1L], byrow = TRUE)
+         as.vector(x)
+     }     else {
+         idxs <- idxs - 1
+         cols <- idxs%/%dim[2L]
+         rows <- idxs%%dim[2L]
+         cols + dim[1L] * rows + 1L
+     }
+ }

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), na_prob = 0) {
+     mode <- match.arg(mode)
+     n <- nrow * ncol
+     if (mode == "logical") {
+         x <- sample(c(FALSE, TRUE), size = n, replace = TRUE)
+     }     else if (mode == "index") {
+         x <- seq_len(n)
+         mode <- "integer"
+     }     else {
+         x <- runif(n, min = range[1], max = range[2])
+     }
+     storage.mode(x) <- mode
+     if (na_prob > 0) 
+         x[sample(n, size = na_prob * n)] <- NA
+     dim(x) <- c(nrow, ncol)
+     x
+ }
> rmatrices <- function(scale = 10, seed = 1, ...) {
+     set.seed(seed)
+     data <- list()
+     data[[1]] <- rmatrix(nrow = scale * 1, ncol = scale * 1, ...)
+     data[[2]] <- rmatrix(nrow = scale * 10, ncol = scale * 10, ...)
+     data[[3]] <- rmatrix(nrow = scale * 100, ncol = scale * 1, ...)
+     data[[4]] <- t(data[[3]])
+     data[[5]] <- rmatrix(nrow = scale * 10, ncol = scale * 100, ...)
+     data[[6]] <- t(data[[5]])
+     names(data) <- sapply(data, FUN = function(x) paste(dim(x), collapse = "x"))
+     data
+ }