matrixStats.benchmarks

---

colCounts() and rowCounts() benchmarks

This report benchmark the performance of colCounts() and rowCounts() against alternative methods.

Alternative methods

• colSums() and rowSums()
• apply() + sum()

Data type “logical”

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 = mode)

Results

10x10 matrix

> X <- data[["10x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.006637	0.0074975	0.0086075	0.0084710	0.0092560	0.021353
1	colCounts	0.007628	0.0084885	0.0101686	0.0099405	0.0106820	0.039195
3	apply+sum	0.042442	0.0444445	0.0475001	0.0453995	0.0470155	0.127609

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.149314	1.132177	1.181375	1.173474	1.154062	1.835573
3	apply+sum	6.394757	5.927909	5.518477	5.359403	5.079462	5.976163

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowSums	0.007533	0.0081670	0.0089452	0.0087740	0.0093810	0.019723
1	rowCounts	0.006886	0.0075145	0.0086889	0.0088570	0.0093275	0.031768
3	apply+sum	0.042136	0.0434245	0.0452873	0.0443105	0.0451515	0.111802

	expr	min	lq	mean	median	uq	max
2	rowSums	1.0000000	1.0000000	1.0000000	1.000000	1.000000	1.000000
1	rowCounts	0.9141112	0.9201053	0.9713412	1.009460	0.994297	1.610708
3	apply+sum	5.5935218	5.3170687	5.0627306	5.050205	4.813080	5.668610

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+10x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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
2	rowCounts	6.886	7.5145	8.68887	8.8570	9.3275	31.768
1	colCounts	7.628	8.4885	10.16864	9.9405	10.6820	39.195

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.107755	1.129616	1.170306	1.122333	1.145216	1.233789

Figure: Benchmarking of colCounts() and rowCounts() on logical+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 matrix

> X <- data[["100x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.036613	0.041205	0.0485826	0.0437540	0.0514525	0.096302
1	colCounts	0.046368	0.054052	0.0619034	0.0569535	0.0664380	0.111596
3	apply+sum	0.279104	0.294825	0.3486344	0.3095320	0.3654120	0.714795

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.266435	1.311783	1.274190	1.301675	1.291249	1.158813
3	apply+sum	7.623085	7.155078	7.176116	7.074370	7.101929	7.422432

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+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	rowCounts	0.057531	0.0645465	0.0710855	0.0679245	0.0790695	0.106873
2	rowSums	0.064276	0.0698320	0.0766227	0.0731645	0.0825285	0.101898
3	apply+sum	0.269382	0.2900325	0.3231464	0.3032355	0.3518315	0.500762

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowSums	1.117241	1.081887	1.077894	1.077144	1.043746	0.9534494
3	apply+sum	4.682380	4.493389	4.545880	4.464302	4.449649	4.6855801

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+100x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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	colCounts	46.368	54.0520	61.90345	56.9535	66.4380	111.596
2	rowCounts	57.531	64.5465	71.08554	67.9245	79.0695	106.873

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCounts	1.240748	1.194156	1.148329	1.192631	1.190125	0.9576777

Figure: Benchmarking of colCounts() and rowCounts() on logical+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 matrix

> X <- data[["1000x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.040137	0.0421630	0.0474956	0.0465065	0.0514625	0.081884
1	colCounts	0.052461	0.0558320	0.0625140	0.0627235	0.0680485	0.097588
3	apply+sum	0.140143	0.1500765	0.1682931	0.1673095	0.1828435	0.253373

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.307048	1.324194	1.316207	1.348704	1.322293	1.191784
3	apply+sum	3.491616	3.559436	3.543342	3.597551	3.552946	3.094292

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+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	rowCounts	0.063240	0.0717505	0.0818468	0.0806795	0.0892285	0.121437
3	apply+sum	0.128911	0.1403345	0.1584276	0.1466985	0.1698245	0.279525
2	rowSums	0.181311	0.1958575	0.2189991	0.2067755	0.2502660	0.291389

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	2.038441	1.955868	1.935661	1.818287	1.903254	2.301811
2	rowSums	2.867030	2.729702	2.675721	2.562925	2.804777	2.399508

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+1000x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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	colCounts	52.461	55.8320	62.51403	62.7235	68.0485	97.588
2	rowCounts	63.240	71.7505	81.84676	80.6795	89.2285	121.437

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	1.205467	1.285114	1.309254	1.286272	1.311249	1.244385

Figure: Benchmarking of colCounts() and rowCounts() on logical+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 matrix

> X <- data[["10x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.030149	0.0337400	0.0392194	0.0360760	0.0420580	0.094415
1	colCounts	0.046557	0.0510365	0.0592526	0.0544325	0.0624885	0.113113
3	apply+sum	1.343127	1.4221290	1.5662842	1.4696620	1.6894220	2.341717

	expr	min	lq	mean	median	uq	max
2	colSums	1.00000	1.000000	1.000000	1.000000	1.00000	1.000000
1	colCounts	1.54423	1.512641	1.510798	1.508829	1.48577	1.198041
3	apply+sum	44.54964	42.149644	39.936445	40.737942	40.16886	24.802383

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowSums	0.045039	0.0469505	0.0558209	0.0488900	0.0635320	0.098466
1	rowCounts	0.049883	0.0529590	0.0629626	0.0565735	0.0701535	0.119705
3	apply+sum	1.301806	1.3719245	1.5317904	1.4278580	1.6298125	2.574432

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.107551	1.127975	1.127939	1.157159	1.104223	1.215699
3	apply+sum	28.903972	29.220658	27.441163	29.205523	25.653411	26.145390

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+10x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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	colCounts	46.557	51.0365	59.25263	54.4325	62.4885	113.113
2	rowCounts	49.883	52.9590	62.96258	56.5735	70.1535	119.705

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	1.071439	1.037669	1.062612	1.039333	1.122663	1.058278

Figure: Benchmarking of colCounts() and rowCounts() on logical+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 matrix

> X <- data[["100x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.261343	0.271250	0.3191512	0.2892330	0.3850325	0.500749
1	colCounts	0.361892	0.366153	0.3924876	0.3792425	0.4025380	0.633757
3	apply+sum	2.108871	2.163654	2.6068709	2.2452695	2.5927860	21.148136

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.384740	1.349873	1.229786	1.311201	1.045465	1.265618
3	apply+sum	8.069361	7.976605	8.168136	7.762840	6.733941	42.233007

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowSums	0.403698	0.4098275	0.4334788	0.4194305	0.4459425	0.592782
1	rowCounts	0.451865	0.4645505	0.4940684	0.4706145	0.4872320	0.791381
3	apply+sum	2.118802	2.1820775	2.5680281	2.2493745	2.4374345	20.257636

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.119314	1.133527	1.139775	1.122032	1.092589	1.335029
3	apply+sum	5.248483	5.324380	5.924230	5.362925	5.465804	34.173838

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+100x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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	colCounts	361.892	366.1530	392.4876	379.2425	402.538	633.757
2	rowCounts	451.865	464.5505	494.0684	470.6145	487.232	791.381

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.0000	1.000000
2	rowCounts	1.248618	1.268733	1.258813	1.240933	1.2104	1.248714

Figure: Benchmarking of colCounts() and rowCounts() on logical+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 matrix

> X <- data[["1000x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on logical+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.264137	0.2669930	0.2846517	0.2701910	0.2808895	0.431141
1	colCounts	0.354495	0.3570615	0.4031536	0.3604420	0.4485740	0.626658
3	apply+sum	0.923800	0.9360290	1.1117882	0.9454405	1.1225460	8.602520

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.342088	1.337344	1.416305	1.334027	1.596977	1.453487
3	apply+sum	3.497428	3.505819	3.905785	3.499156	3.996397	19.952916

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on logical+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	rowCounts	0.458440	0.462624	0.5097221	0.4677710	0.5445670	0.784485
2	rowSums	0.504967	0.506294	0.6187105	0.5093805	0.5350145	8.288021
3	apply+sum	0.927031	0.935527	1.0321958	0.9468965	1.0924870	1.650107

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.0000000	1.000000
2	rowSums	1.101490	1.094396	1.213819	1.088953	0.9824585	10.564920
3	apply+sum	2.022143	2.022219	2.025017	2.024274	2.0061572	2.103427

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on logical+1000x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on logical+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	colCounts	354.495	357.0615	403.1535	360.442	448.574	626.658
2	rowCounts	458.440	462.6240	509.7221	467.771	544.567	784.485

	expr	min	lq	mean	median	uq	max
1	colCounts	1.00000	1.000000	1.000000	1.00000	1.000000	1.000000
2	rowCounts	1.29322	1.295642	1.264337	1.29777	1.213996	1.251855

Figure: Benchmarking of colCounts() and rowCounts() on logical+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Data type “integer”

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 = mode)

Results

10x10 matrix

> X <- data[["10x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.006601	0.0073435	0.0082847	0.0080050	0.0088390	0.020946
1	colCounts	0.007739	0.0084800	0.0097130	0.0091470	0.0103835	0.036537
3	apply+sum	0.040574	0.0431430	0.0451899	0.0439965	0.0449600	0.119376

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.172398	1.154763	1.172396	1.142661	1.174737	1.744343
3	apply+sum	6.146644	5.874992	5.454610	5.496127	5.086548	5.699227

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.006743	0.0074120	0.0085415	0.0085295	0.009106	0.032106
2	rowSums	0.007384	0.0081045	0.0088217	0.0088050	0.009185	0.020475
3	apply+sum	0.040285	0.0422760	0.0446876	0.0434820	0.044954	0.114754

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowSums	1.095061	1.093430	1.032803	1.032300	1.008676	0.6377313
3	apply+sum	5.974344	5.703724	5.231806	5.097837	4.936745	3.5742229

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+10x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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
2	rowCounts	6.743	7.412	8.54153	8.5295	9.1060	32.106
1	colCounts	7.739	8.480	9.71297	9.1470	10.3835	36.537

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.147709	1.144091	1.137146	1.072396	1.140292	1.138012

Figure: Benchmarking of colCounts() and rowCounts() on integer+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 matrix

> X <- data[["100x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.015060	0.0170060	0.0195454	0.0182665	0.0207300	0.060797
2	colSums	0.038388	0.0404445	0.0444467	0.0429670	0.0473505	0.064552
3	apply+sum	0.275794	0.2877540	0.3156357	0.2977740	0.3343565	0.470952

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	2.549004	2.378249	2.274023	2.352229	2.284153	1.061763
3	apply+sum	18.313015	16.920734	16.148849	16.301645	16.129112	7.746303

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.022701	0.0255260	0.0284739	0.0268385	0.0310915	0.056978
2	rowSums	0.066303	0.0706665	0.0770138	0.0742000	0.0847910	0.100826
3	apply+sum	0.265447	0.2830700	0.3139990	0.2967430	0.3425350	0.498593

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	2.920708	2.768413	2.704718	2.764685	2.727144	1.769560
3	apply+sum	11.693185	11.089477	11.027618	11.056616	11.016998	8.750623

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+100x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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	colCounts	15.060	17.006	19.54540	18.2665	20.7300	60.797
2	rowCounts	22.701	25.526	28.47387	26.8385	31.0915	56.978

	expr	min	lq	mean	median	uq	max
1	colCounts	1.00000	1.000	1.000000	1.000000	1.000000	1.0000000
2	rowCounts	1.50737	1.501	1.456807	1.469274	1.499831	0.9371844

Figure: Benchmarking of colCounts() and rowCounts() on integer+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 matrix

> X <- data[["1000x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.014731	0.0162690	0.0188388	0.017977	0.0195425	0.044159
2	colSums	0.039821	0.0416310	0.0452287	0.043642	0.0465020	0.080237
3	apply+sum	0.131626	0.1374625	0.1502718	0.146252	0.1620910	0.219330

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	2.703211	2.558916	2.400825	2.427658	2.379532	1.817002
3	apply+sum	8.935306	8.449352	7.976712	8.135506	8.294282	4.966824

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.021535	0.0241480	0.0273777	0.026082	0.0295095	0.054413
3	apply+sum	0.123974	0.1293700	0.1439065	0.136203	0.1559240	0.217259
2	rowSums	0.186170	0.1939435	0.2148926	0.202771	0.2400175	0.275024

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	5.756861	5.357380	5.256330	5.222107	5.283858	3.992777
2	rowSums	8.644996	8.031452	7.849169	7.774366	8.133567	5.054380

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+1000x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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	colCounts	14.731	16.269	18.83882	17.977	19.5425	44.159
2	rowCounts	21.535	24.148	27.37775	26.082	29.5095	54.413

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	1.461883	1.484295	1.453262	1.450854	1.510017	1.232206

Figure: Benchmarking of colCounts() and rowCounts() on integer+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 matrix

> X <- data[["10x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.014284	0.017018	0.0204368	0.0188795	0.022283	0.054180
2	colSums	0.030019	0.033480	0.0402092	0.0370550	0.043343	0.082655
3	apply+sum	1.311593	1.356021	1.5096524	1.4310310	1.609368	2.192649

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	2.101582	1.967329	1.967491	1.962711	1.945115	1.525563
3	apply+sum	91.822529	79.681602	73.869314	75.798141	72.224050	40.469712

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.018723	0.0206040	0.0244442	0.0223830	0.0264935	0.068516
2	rowSums	0.044882	0.0465705	0.0528485	0.0482725	0.0579840	0.092586
3	apply+sum	1.302213	1.3477245	1.4650412	1.3794150	1.5630885	2.099991

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	2.397159	2.260265	2.162008	2.156659	2.188612	1.351305
3	apply+sum	69.551514	65.410818	59.934201	61.627798	58.998943	30.649644

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+10x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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	colCounts	14.284	17.018	20.43680	18.8795	22.2830	54.180
2	rowCounts	18.723	20.604	24.44416	22.3830	26.4935	68.516

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000
2	rowCounts	1.310767	1.210718	1.196085	1.185572	1.188956	1.2646

Figure: Benchmarking of colCounts() and rowCounts() on integer+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 matrix

> X <- data[["100x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.074836	0.0785405	0.0865373	0.0823900	0.0877020	0.151099
2	colSums	0.261191	0.2710265	0.2952926	0.2802075	0.2941435	0.493963
3	apply+sum	2.051954	2.0882605	2.4662292	2.1891690	2.3201845	20.648812

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	3.490178	3.450786	3.412315	3.400989	3.353897	3.269135
3	apply+sum	27.419344	26.588327	28.499021	26.570810	26.455320	136.657503

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.138607	0.1431090	0.1539760	0.1465505	0.155522	0.269581
2	rowSums	0.403481	0.4090745	0.4296506	0.4197245	0.431219	0.578216
3	apply+sum	2.033093	2.0965345	2.4730421	2.1712950	2.384572	20.292835

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowSums	2.910971	2.858482	2.790374	2.864026	2.77272	2.144869
3	apply+sum	14.668040	14.649914	16.061222	14.816019	15.33270	75.275464

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+100x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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	colCounts	74.836	78.5405	86.53733	82.3900	87.702	151.099
2	rowCounts	138.607	143.1090	153.97596	146.5505	155.522	269.581

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.0000	1.000000
2	rowCounts	1.852143	1.822105	1.779301	1.778741	1.7733	1.784135

Figure: Benchmarking of colCounts() and rowCounts() on integer+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 matrix

> X <- data[["1000x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on integer+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.066060	0.0676605	0.0838680	0.0726380	0.0932175	0.191332
2	colSums	0.263842	0.2671160	0.3009194	0.2715075	0.3320570	0.465800
3	apply+sum	0.853779	0.8656770	1.0871991	0.8954545	1.1339550	8.654829

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
2	colSums	3.993975	3.947887	3.58801	3.737816	3.562174	2.434512
3	apply+sum	12.924296	12.794422	12.96321	12.327632	12.164615	45.234613

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on integer+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	rowCounts	0.145008	0.1473055	0.1653688	0.1527485	0.1824495	0.265969
2	rowSums	0.504530	0.5072465	0.6229228	0.5130110	0.5530105	7.560175
3	apply+sum	0.857034	0.8712995	0.9813462	0.8922660	1.0690230	1.559909

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	3.479325	3.443500	3.766870	3.358534	3.031033	28.425023
3	apply+sum	5.910253	5.914915	5.934289	5.841406	5.859282	5.865003

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on integer+1000x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on integer+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	colCounts	66.060	67.6605	83.86805	72.6380	93.2175	191.332
2	rowCounts	145.008	147.3055	165.36878	152.7485	182.4495	265.969

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	2.195095	2.177127	1.971773	2.102873	1.957245	1.390092

Figure: Benchmarking of colCounts() and rowCounts() on integer+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Data type “double”

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 = mode)

Results

10x10 matrix

> X <- data[["10x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colSums	0.006350	0.0072695	0.0085632	0.0082510	0.0088780	0.021279
1	colCounts	0.007738	0.0085720	0.0100339	0.0094595	0.0103480	0.037644
3	apply+sum	0.042346	0.0442090	0.0473068	0.0449065	0.0460955	0.119980

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.218583	1.179173	1.171743	1.146467	1.165578	1.769068
3	apply+sum	6.668661	6.081436	5.524435	5.442552	5.192104	5.638423

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.006733	0.0074745	0.0087072	0.0088385	0.0091630	0.033580
2	rowSums	0.007404	0.0080240	0.0089398	0.0088565	0.0094105	0.019765
3	apply+sum	0.042170	0.0435875	0.0458676	0.0442790	0.0449420	0.167430

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowSums	1.099658	1.073517	1.026719	1.002036	1.027011	0.5885944
3	apply+sum	6.263181	5.831494	5.267778	5.009787	4.904725	4.9860036

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+10x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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
2	rowCounts	6.733	7.4745	8.70720	8.8385	9.163	33.580
1	colCounts	7.738	8.5720	10.03387	9.4595	10.348	37.644

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.149265	1.146833	1.152365	1.070261	1.129325	1.121024

Figure: Benchmarking of colCounts() and rowCounts() on double+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 matrix

> X <- data[["100x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.019315	0.0216215	0.0248835	0.0229120	0.0264315	0.068351
2	colSums	0.035221	0.0370065	0.0418328	0.0406325	0.0451275	0.061254
3	apply+sum	0.274895	0.2825740	0.3177964	0.2952570	0.3340440	0.477387

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
2	colSums	1.823505	1.71156	1.681148	1.773416	1.707338	0.8961683
3	apply+sum	14.232203	13.06912	12.771369	12.886566	12.638102	6.9843455

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.018674	0.0212115	0.0234508	0.0222250	0.0250200	0.050941
2	rowSums	0.061601	0.0656455	0.0717409	0.0687150	0.0769245	0.097380
3	apply+sum	0.260498	0.2779735	0.3063011	0.2888495	0.3302950	0.503861

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowSums	3.298758	3.094807	3.059209	3.091788	3.07452	1.911623
3	apply+sum	13.949770	13.104849	13.061436	12.996603	13.20124	9.891070

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+100x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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
2	rowCounts	18.674	21.2115	23.4508	22.225	25.0200	50.941
1	colCounts	19.315	21.6215	24.8835	22.912	26.4315	68.351

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.034326	1.019329	1.061094	1.030911	1.056415	1.341768

Figure: Benchmarking of colCounts() and rowCounts() on double+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 matrix

> X <- data[["1000x10"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.019521	0.0220475	0.0249289	0.0239215	0.0258260	0.074930
2	colSums	0.036529	0.0398005	0.0452394	0.0427700	0.0461280	0.142064
3	apply+sum	0.134231	0.1436985	0.1600036	0.1560095	0.1688235	0.299705

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.871267	1.805216	1.814739	1.787931	1.786107	1.895956
3	apply+sum	6.876236	6.517678	6.418408	6.521727	6.536959	3.999800

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.019320	0.0218085	0.0263440	0.0251540	0.026535	0.071538
3	apply+sum	0.129371	0.1352715	0.1611321	0.1422125	0.168951	0.362152
2	rowSums	0.190320	0.1977460	0.2210332	0.2114500	0.246531	0.301963

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	6.696222	6.202696	6.116461	5.653673	6.367100	5.062372
2	rowSums	9.850932	9.067382	8.390265	8.406218	9.290786	4.221015

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+1000x10 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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	colCounts	19.521	22.0475	24.92886	23.9215	25.826	74.930
2	rowCounts	19.320	21.8085	26.34401	25.1540	26.535	71.538

	expr	min	lq	mean	median	uq	max
1	colCounts	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.0000000
2	rowCounts	0.9897034	0.9891598	1.056768	1.051523	1.027453	0.9547311

Figure: Benchmarking of colCounts() and rowCounts() on double+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 matrix

> X <- data[["10x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.016777	0.0189145	0.0248176	0.0221425	0.026590	0.074802
2	colSums	0.027659	0.0317780	0.0377182	0.0353500	0.038784	0.092343
3	apply+sum	1.287584	1.4017560	1.5128844	1.4757060	1.573189	2.590975

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.648626	1.680087	1.519816	1.596477	1.458594	1.234499
3	apply+sum	76.746975	74.110127	60.960066	66.645862	59.164686	34.637777

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.015787	0.0190205	0.0224049	0.0207755	0.023704	0.065639
2	rowSums	0.042511	0.0468050	0.0519133	0.0489985	0.054957	0.092636
3	apply+sum	1.295856	1.3844675	1.4861423	1.4685930	1.555566	1.830412

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	2.692785	2.460766	2.317049	2.358475	2.318469	1.411295
3	apply+sum	82.083740	72.788176	66.331071	70.688696	65.624620	27.886043

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+10x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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
2	rowCounts	15.787	19.0205	22.40492	20.7755	23.704	65.639
1	colCounts	16.777	18.9145	24.81763	22.1425	26.590	74.802

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.00000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.06271	0.9944271	1.107687	1.065799	1.121752	1.139597

Figure: Benchmarking of colCounts() and rowCounts() on double+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 matrix

> X <- data[["100x1000"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.106208	0.110792	0.1249620	0.1196105	0.1286505	0.211605
2	colSums	0.234996	0.244717	0.2731302	0.2557090	0.2808200	0.488140
3	apply+sum	2.017583	2.098267	2.5172652	2.1463755	2.5706040	20.252990

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	2.212602	2.208797	2.185705	2.137847	2.182813	2.306845
3	apply+sum	18.996526	18.938795	20.144237	17.944708	19.981298	95.711302

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.111257	0.120393	0.1369497	0.1331260	0.1457465	0.240432
2	rowSums	0.378523	0.400822	0.4221277	0.4146545	0.4305350	0.587751
3	apply+sum	2.064423	2.199650	2.5900267	2.2840775	2.5770720	19.405890

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.00000	1.00000	1.000000	1.000000	1.000000	1.000000
2	rowSums	3.40224	3.32928	3.082356	3.114752	2.953999	2.444562
3	apply+sum	18.55544	18.27058	18.912250	17.157261	17.681879	80.712592

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+100x1000 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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	colCounts	106.208	110.792	124.9621	119.6105	128.6505	211.605
2	rowCounts	111.257	120.393	136.9497	133.1260	145.7465	240.432

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.00000	1.000000	1.000000	1.00000
2	rowCounts	1.047539	1.086658	1.09593	1.112996	1.132887	1.13623

Figure: Benchmarking of colCounts() and rowCounts() on double+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 matrix

> X <- data[["1000x100"]]
> value <- 42

> colStats <- microbenchmark(colCounts = colCounts(X, value = value, na.rm = FALSE), colSums = colSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 2L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

> X <- t(X)

> rowStats <- microbenchmark(rowCounts = rowCounts(X, value = value, na.rm = FALSE), rowSums = rowSums(X == 
+     value, na.rm = FALSE), `apply+sum` = apply(X, MARGIN = 1L, FUN = function(x) sum(x == value, 
+     na.rm = FALSE)), unit = "ms")

Table: Benchmarking of colCounts(), colSums() and apply+sum() on double+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCounts	0.100316	0.1228340	0.1337999	0.1307415	0.144953	0.219630
2	colSums	0.239877	0.2684070	0.3544476	0.2955835	0.316224	6.538396
3	apply+sum	0.869230	0.9984675	1.0835718	1.1017920	1.132891	1.663588

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	2.391214	2.185120	2.649087	2.260824	2.181562	29.770050
3	apply+sum	8.664919	8.128592	8.098448	8.427255	7.815571	7.574503

Table: Benchmarking of rowCounts(), rowSums() and apply+sum() on double+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	rowCounts	0.117644	0.1241655	0.1348207	0.1272840	0.1397885	0.225083
2	rowSums	0.478986	0.4848400	0.5212520	0.5029105	0.5274120	0.733624
3	apply+sum	0.880771	0.9132125	1.0615535	0.9532880	1.0444710	7.144646

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	4.071487	3.904788	3.866261	3.951090	3.772928	3.259349
3	apply+sum	7.486748	7.354801	7.873817	7.489457	7.471795	31.742273

Figure: Benchmarking of colCounts(), colSums() and apply+sum() on double+1000x100 data as well as rowCounts(), rowSums() and apply+sum() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCounts() and rowCounts() on double+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
2	rowCounts	117.644	124.1655	134.8207	127.2840	139.7885	225.083
1	colCounts	100.316	122.8340	133.7999	130.7415	144.9530	219.630

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.0000000	1.0000000	1.0000000	1.000000	1.000000	1.0000000
1	colCounts	0.8527082	0.9892764	0.9924287	1.027164	1.036945	0.9757734

Figure: Benchmarking of colCounts() and rowCounts() on double+1000x100 data (original and transposed). 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         
[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 38.03 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Henrik Bengtsson. Last updated on 2021-08-25 22:12:40 (+0200 UTC). Powered by RSP.