colRowProds

matrixStats: Benchmark report

---

colProds() and rowProds() benchmarks

This report benchmark the performance of colProds() and rowProds() against alternative methods.

Alternative methods

• colProds()/rowProds() using method="expSumLog"
• apply() + prod()
• apply() + product()

Data

> 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
+ }
> data <- rmatrices(mode = "double")

Results

10x10 matrix

> X <- data[["10x10"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	0.008800	0.0094145	0.0099799	0.0098050	0.0102595	0.021529
2	colProds w/ expSumLog	0.014226	0.0148440	0.0154650	0.0155405	0.0158780	0.019312
3	apply+prod	0.020025	0.0210960	0.0215453	0.0214540	0.0218900	0.028227
4	apply+product	0.027252	0.0277515	0.0293899	0.0280850	0.0284865	0.123770

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colProds w/ expSumLog	1.616591	1.576717	1.549612	1.584957	1.547639	0.8970226
3	apply+prod	2.275568	2.240799	2.158871	2.188067	2.133632	1.3111152
4	apply+product	3.096818	2.947740	2.944905	2.864355	2.776597	5.7489897

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	0.008971	0.0094915	0.0101705	0.0100415	0.0104500	0.020142
2	rowProds w/ expSumLog	0.014368	0.0149075	0.0159106	0.0156585	0.0162505	0.041111
3	apply+prod	0.019603	0.0206265	0.0223654	0.0212315	0.0216405	0.103849
4	apply+product	0.026428	0.0273865	0.0279480	0.0276900	0.0281740	0.037545

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowProds w/ expSumLog	1.601605	1.570616	1.564385	1.559379	1.555072	2.041059
3	apply+prod	2.185152	2.173155	2.199039	2.114375	2.070861	5.155843
4	apply+product	2.945937	2.885371	2.747940	2.757556	2.696077	1.864016

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 10x10 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	8.800	9.4145	9.97990	9.8050	10.2595	21.529
2	rowProds w/ direct	8.971	9.4915	10.17054	10.0415	10.4500	20.142

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.00000	1.000000	1.0000000
2	rowProds w/ direct	1.019432	1.008179	1.019102	1.02412	1.018568	0.9355753

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 matrix

> X <- data[["100x100"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	0.114208	0.1154645	0.1184836	0.1166710	0.1191515	0.168758
3	apply+prod	0.185868	0.1880695	0.1918381	0.1896675	0.1931980	0.212631
2	colProds w/ expSumLog	0.362362	0.3652075	0.3701785	0.3678675	0.3719820	0.397963
4	apply+product	0.444140	0.4475500	0.4544683	0.4501985	0.4545590	0.594481

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	1.627452	1.628808	1.619111	1.625661	1.621448	1.259976
2	colProds w/ expSumLog	3.172825	3.162942	3.124301	3.153033	3.121925	2.358188
4	apply+product	3.888869	3.876083	3.835705	3.858701	3.814967	3.522683

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	0.115724	0.1179720	0.1204872	0.1190215	0.1209745	0.167360
3	apply+prod	0.144730	0.1471305	0.1543799	0.1487155	0.1518415	0.540496
2	rowProds w/ expSumLog	0.366354	0.3695335	0.3764591	0.3714560	0.3754380	0.508744
4	apply+product	0.402775	0.4066835	0.4141205	0.4104350	0.4154095	0.519079

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	1.250648	1.247165	1.281297	1.249484	1.255153	3.229541
2	rowProds w/ expSumLog	3.165757	3.132383	3.124474	3.120915	3.103447	3.039818
4	apply+product	3.480479	3.447288	3.437050	3.448411	3.433860	3.101571

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 100x100 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	114.208	115.4645	118.4836	116.6710	119.1515	168.758
2	rowProds w/ direct	115.724	117.9720	120.4872	119.0215	120.9745	167.360

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.00000	1.000000	1.0000	1.0000000
2	rowProds w/ direct	1.013274	1.021717	1.01691	1.020146	1.0153	0.9917159

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 matrix

> X <- data[["1000x10"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	0.056584	0.0594655	0.0608798	0.0605925	0.0617985	0.076076
3	apply+prod	0.118388	0.1214215	0.1260853	0.1234645	0.1250075	0.223058
2	colProds w/ expSumLog	0.264604	0.2671600	0.2710972	0.2685945	0.2701615	0.351917
4	apply+product	0.328278	0.3319470	0.3385626	0.3333385	0.3350340	0.719578

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	2.092252	2.041881	2.071054	2.037620	2.022824	2.932042
2	colProds w/ expSumLog	4.676304	4.492689	4.452994	4.432801	4.371651	4.625861
4	apply+product	5.801605	5.582178	5.561168	5.501316	5.421394	9.458673

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 1000x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	0.058456	0.0606645	0.0625605	0.0625085	0.0636825	0.072641
3	apply+prod	0.076469	0.0810880	0.0838753	0.0829305	0.0848060	0.106722
2	rowProds w/ expSumLog	0.267670	0.2705885	0.2730204	0.2723735	0.2736355	0.292592
4	apply+product	0.287905	0.2904395	0.2956810	0.2922185	0.2960595	0.390372

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	1.308146	1.336663	1.340706	1.326708	1.331700	1.469170
2	rowProds w/ expSumLog	4.579000	4.460409	4.364101	4.357383	4.296871	4.027918
4	apply+product	4.925157	4.787635	4.726320	4.674860	4.648993	5.373990

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 1000x10 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 1000x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	56.584	59.4655	60.87977	60.5925	61.7985	76.076
2	rowProds w/ direct	58.456	60.6645	62.56052	62.5085	63.6825	72.641

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowProds w/ direct	1.033084	1.020163	1.027608	1.031621	1.030486	0.9548478

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 matrix

> X <- data[["10x1000"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	0.644066	0.684814	0.7599742	0.6989315	0.7216265	5.798954
3	apply+prod	0.812111	0.834671	0.8740589	0.8537895	0.8796950	1.472521
2	colProds w/ expSumLog	1.226847	1.278935	1.3722292	1.3015150	1.3495120	6.507626
4	apply+product	1.464395	1.505959	1.5636033	1.5311550	1.5869835	2.282037

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	apply+prod	1.260913	1.218829	1.150116	1.221564	1.219045	0.2539287
2	colProds w/ expSumLog	1.904847	1.867565	1.805626	1.862150	1.870098	1.1222069
4	apply+product	2.273672	2.199078	2.057443	2.190708	2.199176	0.3935256

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	0.665562	0.702727	0.7232769	0.715927	0.7426305	0.802537
3	apply+prod	0.798166	0.830586	0.8740196	0.858047	0.8960415	1.202002
2	rowProds w/ expSumLog	1.248327	1.308507	1.4198041	1.349049	1.3859660	6.681312
4	apply+product	1.466300	1.523035	1.6262461	1.564934	1.6061500	6.903179

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	1.199236	1.181947	1.208416	1.198512	1.206578	1.497753
2	rowProds w/ expSumLog	1.875598	1.862042	1.963016	1.884339	1.866293	8.325239
4	apply+product	2.203101	2.167321	2.248442	2.185886	2.162785	8.601696

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 10x1000 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	644.066	684.814	759.9742	698.9315	721.6265	5798.954
2	rowProds w/ direct	665.562	702.727	723.2769	715.9270	742.6305	802.537

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowProds w/ direct	1.033376	1.026158	0.9517124	1.024316	1.029106	0.1383934

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 matrix

> X <- data[["100x1000"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.109562	1.120069	1.179608	1.139917	1.175838	1.720146
3	apply+prod	1.734878	1.757721	2.148371	1.782883	1.842180	17.418038
2	colProds w/ expSumLog	3.583810	3.615213	4.044786	3.647726	3.716857	19.286408
4	apply+product	4.305582	4.347441	4.700679	4.420883	4.504060	21.477891

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
3	apply+prod	1.563570	1.569297	1.821258	1.564046	1.566695	10.12591
2	colProds w/ expSumLog	3.229932	3.227670	3.428923	3.199993	3.161028	11.21208
4	apply+product	3.880434	3.881404	3.984949	3.878250	3.830510	12.48609

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.150689	1.169666	1.258584	1.179300	1.222171	1.889199
3	apply+prod	1.319024	1.338129	1.866213	1.355963	1.388661	17.378289
2	rowProds w/ expSumLog	3.640529	3.670341	3.997877	3.680613	3.730546	20.434404
4	apply+product	3.881884	3.928486	4.268454	3.944160	4.031906	20.527597

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
3	apply+prod	1.146291	1.144027	1.482788	1.149803	1.136224	9.19876
2	rowProds w/ expSumLog	3.163782	3.137939	3.176489	3.121015	3.052392	10.81644
4	apply+product	3.373530	3.358639	3.391473	3.344492	3.298970	10.86577

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 100x1000 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 100x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.109562	1.120069	1.179608	1.139917	1.175838	1.720146
2	rowProds w/ direct	1.150689	1.169666	1.258584	1.179300	1.222171	1.889199

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.00000	1.000000	1.000000	1.000000	1.000000
2	rowProds w/ direct	1.037066	1.04428	1.066951	1.034549	1.039405	1.098278

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 matrix

> X <- data[["1000x100"]]
> colStats <- microbenchmark(`colProds w/ direct` = colProds(X, method = "direct", na.rm = FALSE), 
+     `colProds w/ expSumLog` = colProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 2L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 2L, FUN = product, 
+         na.rm = FALSE), unit = "ms")
> X <- t(X)
> rowStats <- microbenchmark(`rowProds w/ direct` = rowProds(X, method = "direct", na.rm = FALSE), 
+     `rowProds w/ expSumLog` = rowProds(X, method = "expSumLog", na.rm = FALSE), `apply+prod` = apply(X, 
+         MARGIN = 1L, FUN = prod, na.rm = FALSE), `apply+product` = apply(X, MARGIN = 1L, FUN = product, 
+         na.rm = FALSE), unit = "ms")

Table: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	0.363557	0.3716730	0.3887810	0.3839445	0.3921995	0.477998
3	apply+prod	0.640136	0.6542385	0.9097683	0.6858240	0.7200375	7.951854
2	colProds w/ expSumLog	2.418585	2.4573250	2.5319060	2.4730465	2.4955020	3.160826
4	apply+product	2.732607	2.7511875	2.9202926	2.7832505	2.8313660	9.714860

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+prod	1.760758	1.760253	2.340053	1.786258	1.835896	16.635747
2	colProds w/ expSumLog	6.652561	6.611524	6.512422	6.441156	6.362838	6.612634
4	apply+product	7.516310	7.402172	7.511408	7.249096	7.219198	20.324060

Table: Benchmarking of rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on 1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	0.594052	0.602855	0.7037156	0.609530	0.6415405	7.690274
3	apply+prod	0.666871	0.684659	0.7341861	0.715175	0.7453200	1.080402
2	rowProds w/ expSumLog	2.701753	2.731766	2.9677295	2.741329	2.7817565	10.744616
4	apply+product	2.767487	2.815783	3.0263775	2.830687	2.8881060	15.105563

	expr	min	lq	mean	median	uq	max
1	rowProds w/ direct	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	apply+prod	1.122580	1.135694	1.043300	1.173322	1.161766	0.1404894
2	rowProds w/ expSumLog	4.548008	4.531382	4.217229	4.497448	4.336057	1.3971695
4	apply+product	4.658661	4.670746	4.300569	4.644050	4.501830	1.9642425

Figure: Benchmarking of colProds w/ direct(), colProds w/ expSumLog(), apply+prod() and apply+product() on 1000x100 data as well as rowProds w/ direct(), rowProds w/ expSumLog(), apply+prod() and apply+product() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 1000x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	363.557	371.673	388.7810	383.9445	392.1995	477.998
2	rowProds w/ direct	594.052	602.855	703.7156	609.5300	641.5405	7690.274

	expr	min	lq	mean	median	uq	max
1	colProds w/ direct	1.000	1.000000	1.000000	1.000000	1.00000	1.00000
2	rowProds w/ direct	1.634	1.622004	1.810057	1.587547	1.63575	16.08851

Figure: Benchmarking of colProds w/ direct() and rowProds w/ direct() on 1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

R version 3.6.1 Patched (2019-08-27 r77078)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.3 LTS

Matrix products: default
BLAS:   /home/hb/software/R-devel/R-3-6-branch/lib/R/lib/libRblas.so
LAPACK: /home/hb/software/R-devel/R-3-6-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_1.4-6    matrixStats_0.55.0-9000 ggplot2_3.2.1          
[4] knitr_1.24              R.devices_2.16.0        R.utils_2.9.0          
[7] R.oo_1.22.0             R.methodsS3_1.7.1       history_0.0.0-9002     

loaded via a namespace (and not attached):
 [1] Biobase_2.45.0       bit64_0.9-7          splines_3.6.1       
 [4] network_1.15         assertthat_0.2.1     highr_0.8           
 [7] stats4_3.6.1         blob_1.2.0           robustbase_0.93-5   
[10] pillar_1.4.2         RSQLite_2.1.2        backports_1.1.4     
[13] lattice_0.20-38      glue_1.3.1           digest_0.6.20       
[16] colorspace_1.4-1     sandwich_2.5-1       Matrix_1.2-17       
[19] XML_3.98-1.20        lpSolve_5.6.13.3     pkgconfig_2.0.2     
[22] genefilter_1.66.0    purrr_0.3.2          ergm_3.10.4         
[25] xtable_1.8-4         mvtnorm_1.0-11       scales_1.0.0        
[28] tibble_2.1.3         annotate_1.62.0      IRanges_2.18.2      
[31] TH.data_1.0-10       withr_2.1.2          BiocGenerics_0.30.0 
[34] lazyeval_0.2.2       mime_0.7             survival_2.44-1.1   
[37] magrittr_1.5         crayon_1.3.4         statnet.common_4.3.0
[40] memoise_1.1.0        laeken_0.5.0         R.cache_0.13.0      
[43] MASS_7.3-51.4        R.rsp_0.43.1         tools_3.6.1         
[46] multcomp_1.4-10      S4Vectors_0.22.1     trust_0.1-7         
[49] munsell_0.5.0        AnnotationDbi_1.46.1 compiler_3.6.1      
[52] rlang_0.4.0          grid_3.6.1           RCurl_1.95-4.12     
[55] cwhmisc_6.6          rappdirs_0.3.1       labeling_0.3        
[58] bitops_1.0-6         base64enc_0.1-3      boot_1.3-23         
[61] gtable_0.3.0         codetools_0.2-16     DBI_1.0.0           
[64] markdown_1.1         R6_2.4.0             zoo_1.8-6           
[67] dplyr_0.8.3          bit_1.1-14           zeallot_0.1.0       
[70] parallel_3.6.1       Rcpp_1.0.2           vctrs_0.2.0         
[73] DEoptimR_1.0-8       tidyselect_0.2.5     xfun_0.9            
[76] coda_0.19-3         

Total processing time was 17.46 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Henrik Bengtsson. Last updated on 2019-09-10 20:49:13 (-0700 UTC). Powered by RSP.

<script> var link = document.createElement('link'); link.rel = 'icon'; link.href = "data:image/png;base64,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" document.getElementsByTagName('head')[0].appendChild(link); </script>