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	0.006501	0.0073220	0.0083084	0.0081615	0.0088265	0.027202
1	colCounts	0.008136	0.0089050	0.0104627	0.0097525	0.0110085	0.048968
3	apply+sum	0.040800	0.0431555	0.0452735	0.0439785	0.0452470	0.115865

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.251500	1.216198	1.259291	1.194940	1.247210	1.800162
3	apply+sum	6.275957	5.893950	5.449132	5.388532	5.126267	4.259430

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	0.007394	0.0080365	0.0088441	0.0087190	0.009276	0.019050
1	rowCounts	0.007041	0.0076265	0.0090828	0.0092620	0.009751	0.034029
3	apply+sum	0.041004	0.0426325	0.0455665	0.0433915	0.044400	0.112803

	expr	min	lq	mean	median	uq	max
2	rowSums	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	0.9522586	0.9489828	1.026994	1.062278	1.051207	1.786299
3	apply+sum	5.5455775	5.3048591	5.152195	4.976660	4.786546	5.921417

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	7.041	7.6265	9.08284	9.2620	9.7510	34.029
1	colCounts	8.136	8.9050	10.46268	9.7525	11.0085	48.968

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.155518	1.167639	1.151917	1.052958	1.128961	1.439008

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	0.036407	0.0398030	0.0440769	0.0424175	0.0465420	0.070074
1	colCounts	0.076051	0.0824960	0.0916281	0.0872825	0.1008040	0.160833
3	apply+sum	0.269949	0.2900645	0.3206545	0.3039580	0.3385325	0.490250

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	2.088912	2.072608	2.078826	2.057700	2.165872	2.295188
3	apply+sum	7.414755	7.287503	7.274895	7.165863	7.273699	6.996175

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
2	rowSums	0.062661	0.0684775	0.0761178	0.0720920	0.0824805	0.115206
1	rowCounts	0.070389	0.0795345	0.0875144	0.0834775	0.0965375	0.132294
3	apply+sum	0.262021	0.2856800	0.3240388	0.3087325	0.3536190	0.539589

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.123330	1.161469	1.149722	1.157930	1.170428	1.148326
3	apply+sum	4.181564	4.171881	4.257068	4.282479	4.287304	4.683688

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
2	rowCounts	70.389	79.5345	87.51437	83.4775	96.5375	132.294
1	colCounts	76.051	82.4960	91.62811	87.2825	100.8040	160.833

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.080439	1.037235	1.047006	1.045581	1.044195	1.215724

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	0.039955	0.0415010	0.0463280	0.0447725	0.0496180	0.113803
1	colCounts	0.081870	0.0857645	0.0964906	0.0953415	0.1049435	0.151107
3	apply+sum	0.141418	0.1455040	0.1668182	0.1626485	0.1817585	0.278370

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	2.049055	2.066565	2.082772	2.129466	2.115029	1.327794
3	apply+sum	3.539432	3.506036	3.600811	3.632777	3.663156	2.446069

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	0.070871	0.0861760	0.0941584	0.091879	0.1038375	0.141825
3	apply+sum	0.122591	0.1336110	0.1493851	0.144210	0.1591510	0.244967
2	rowSums	0.170026	0.1906035	0.2066102	0.199714	0.2238580	0.280389

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	1.729777	1.550443	1.586530	1.569564	1.532693	1.727248
2	rowSums	2.399091	2.211793	2.194284	2.173663	2.155849	1.977007

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
2	rowCounts	70.871	86.1760	94.15836	91.8790	103.8375	141.825
1	colCounts	81.870	85.7645	96.49059	95.3415	104.9435	151.107

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.155198	0.9952249	1.024769	1.037685	1.010651	1.065447

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	0.030777	0.033685	0.0380735	0.036006	0.039692	0.086183
1	colCounts	0.072744	0.078724	0.0877118	0.083235	0.088893	0.153550
3	apply+sum	1.352990	1.447261	1.5418828	1.506645	1.608526	2.154248

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
1	colCounts	2.363583	2.337064	2.303748	2.311698	2.23957	1.781674
3	apply+sum	43.961075	42.964554	40.497512	41.844276	40.52521	24.996206

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	0.045650	0.0505585	0.0563910	0.0534225	0.0594085	0.100107
1	rowCounts	0.062051	0.0702600	0.0813624	0.0757050	0.0881570	0.141772
3	apply+sum	1.344391	1.4742460	1.5905001	1.5300320	1.6568185	2.176518

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.359277	1.389677	1.442827	1.417099	1.483912	1.416205
3	apply+sum	29.449967	29.159212	28.204862	28.640217	27.888577	21.741916

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
2	rowCounts	62.051	70.260	81.36245	75.705	88.157	141.772
1	colCounts	72.744	78.724	87.71178	83.235	88.893	153.550

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.172326	1.120467	1.078038	1.099465	1.008349	1.083077

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	0.261205	0.2856225	0.4783527	0.2945520	0.3105435	17.695614
1	colCounts	0.600921	0.6491945	0.6698462	0.6678425	0.6782875	0.925825
3	apply+sum	2.150766	2.3229135	2.4447240	2.4007440	2.4493115	3.797806

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colCounts	2.300572	2.272911	1.400319	2.267316	2.184195	0.0523195
3	apply+sum	8.234015	8.132810	5.110714	8.150493	7.887177	0.2146185

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	0.405189	0.438964	0.4550076	0.4535765	0.4643900	0.556336
1	rowCounts	0.572433	0.628539	0.6556373	0.6439620	0.6534015	0.984505
3	apply+sum	2.151278	2.330856	2.6260043	2.3835415	2.4454930	19.961024

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.412756	1.431869	1.440937	1.419743	1.407010	1.769623
3	apply+sum	5.309320	5.309902	5.771342	5.254993	5.266033	35.879440

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
2	rowCounts	572.433	628.5390	655.6373	643.9620	653.4015	984.505
1	colCounts	600.921	649.1945	669.8462	667.8425	678.2875	925.825

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colCounts	1.049767	1.032863	1.021672	1.037084	1.038087	0.9403964

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	0.264316	0.2752995	0.2935237	0.2865115	0.2981235	0.401451
1	colCounts	0.592004	0.6237240	0.6820136	0.6446900	0.7149680	1.010604
3	apply+sum	0.932985	0.9867185	1.1448083	1.0285045	1.1287550	8.547166

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	2.239759	2.265620	2.323538	2.250137	2.398228	2.517378
3	apply+sum	3.529809	3.584164	3.900224	3.589749	3.786199	21.290683

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
2	rowSums	0.505743	0.5476215	0.5771170	0.5536975	0.5746035	0.922504
1	rowCounts	0.578038	0.6270185	0.6690668	0.6403885	0.6807910	1.010234
3	apply+sum	0.940992	1.0089565	1.1637090	1.0268850	1.1337085	8.836396

	expr	min	lq	mean	median	uq	max
2	rowSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	1.142948	1.144985	1.159326	1.156567	1.184801	1.095100
3	apply+sum	1.860613	1.842434	2.016418	1.854596	1.973027	9.578708

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
2	rowCounts	578.038	627.0185	669.0668	640.3885	680.791	1010.234
1	colCounts	592.004	623.7240	682.0136	644.6900	714.968	1010.604

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.024161	0.9947458	1.019351	1.006717	1.050202	1.000366

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	0.006626	0.007294	0.0085629	0.0082660	0.0089155	0.021699
1	colCounts	0.008518	0.008998	0.0105554	0.0099655	0.0112540	0.038060
3	apply+sum	0.041677	0.043320	0.0453861	0.0442305	0.0452055	0.117208

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.285542	1.233617	1.232699	1.205601	1.262296	1.753998
3	apply+sum	6.289918	5.939128	5.300329	5.350895	5.070439	5.401539

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
2	rowSums	0.007367	0.0082515	0.0090520	0.0089205	0.0095640	0.020015
1	rowCounts	0.007200	0.0077835	0.0093122	0.0093590	0.0098685	0.034042
3	apply+sum	0.041930	0.0432210	0.0460938	0.0442305	0.0465350	0.120707

	expr	min	lq	mean	median	uq	max
2	rowSums	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	0.9773313	0.943283	1.028749	1.049156	1.031838	1.700824
3	apply+sum	5.6915977	5.237957	5.092116	4.958298	4.865642	6.030827

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	7.200	7.7835	9.31224	9.3590	9.8685	34.042
1	colCounts	8.518	8.9980	10.55545	9.9655	11.2540	38.060

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.183056	1.156035	1.133503	1.064804	1.140396	1.118031

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
2	colSums	0.037768	0.0405210	0.0458021	0.0444525	0.0496990	0.093093
1	colCounts	0.042724	0.0475290	0.0533364	0.0509165	0.0578555	0.109744
3	apply+sum	0.264001	0.2859985	0.3263118	0.3046985	0.3565860	0.593220

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.131222	1.172947	1.164496	1.145414	1.164118	1.178864
3	apply+sum	6.990071	7.058032	7.124382	6.854474	7.174913	6.372337

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	0.031985	0.0356685	0.0399702	0.0377125	0.0435030	0.070805
2	rowSums	0.064445	0.0692020	0.0764049	0.0726110	0.0820465	0.119230
3	apply+sum	0.258466	0.2754525	0.3094303	0.2902240	0.3294460	0.483973

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	2.014851	1.940143	1.911548	1.925383	1.885996	1.683921
3	apply+sum	8.080850	7.722570	7.741530	7.695698	7.572949	6.835294

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
2	rowCounts	31.985	35.6685	39.97017	37.7125	43.5030	70.805
1	colCounts	42.724	47.5290	53.33639	50.9165	57.8555	109.744

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.00000	1.000000	1.000000	1.00000	1.000000
1	colCounts	1.335751	1.33252	1.334405	1.350123	1.32992	1.549947

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
2	colSums	0.039845	0.041388	0.0454423	0.0444495	0.0479125	0.083981
1	colCounts	0.044569	0.046770	0.0520864	0.0513160	0.0557325	0.082555
3	apply+sum	0.130870	0.135431	0.1509504	0.1474645	0.1664075	0.224711

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.118559	1.130038	1.146211	1.154479	1.163214	0.983020
3	apply+sum	3.284477	3.272229	3.321806	3.317574	3.473154	2.675736

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	0.033575	0.0375830	0.0418534	0.0412785	0.0456020	0.073876
3	apply+sum	0.120425	0.1293015	0.1431415	0.1362375	0.1510665	0.224308
2	rowSums	0.180225	0.1928710	0.2117158	0.2003005	0.2311610	0.274867

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	3.586746	3.440425	3.420066	3.300447	3.312717	3.036277
2	rowSums	5.367833	5.131868	5.058504	4.852417	5.069098	3.720654

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
2	rowCounts	33.575	37.583	41.85343	41.2785	45.6020	73.876
1	colCounts	44.569	46.770	52.08644	51.3160	55.7325	82.555

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
1	colCounts	1.327446	1.244446	1.244496	1.243165	1.22215	1.117481

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
2	colSums	0.030778	0.034657	0.0405747	0.0369215	0.0429595	0.084831
1	colCounts	0.038625	0.043015	0.0496964	0.0468140	0.0522395	0.095721
3	apply+sum	1.310625	1.412213	1.5375623	1.4823900	1.5930885	2.639316

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.254955	1.241163	1.224811	1.267933	1.216017	1.128373
3	apply+sum	42.583176	40.748276	37.894589	40.149777	37.083497	31.112636

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	0.027429	0.0308805	0.0356957	0.0335800	0.0377655	0.088183
2	rowSums	0.045477	0.0498400	0.0552398	0.0522535	0.0581305	0.094048
3	apply+sum	1.300316	1.4117940	1.5084055	1.4651420	1.5634850	2.087419

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.00000	1.000000	1.000000	1.00000	1.000000	1.000000
2	rowSums	1.65799	1.613964	1.547518	1.55609	1.539249	1.066509
3	apply+sum	47.40661	45.717977	42.257298	43.63139	41.399823	23.671445

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
2	rowCounts	27.429	30.8805	35.69574	33.580	37.7655	88.183
1	colCounts	38.625	43.0150	49.69637	46.814	52.2395	95.721

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.00000	1.000000	1.000000	1.00000	1.000000
1	colCounts	1.408181	1.39295	1.392221	1.394104	1.38326	1.085481

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
2	colSums	0.264572	0.2920820	0.4882184	0.3013125	0.312993	17.844516
1	colCounts	0.307298	0.3376245	0.3505225	0.3475470	0.355038	0.528144
3	apply+sum	2.056618	2.2697440	2.3934044	2.3261405	2.393671	3.689791

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
1	colCounts	1.161491	1.155924	0.7179624	1.153444	1.134332	0.0295970
3	apply+sum	7.773377	7.770914	4.9023232	7.720027	7.647684	0.2067745

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	0.227010	0.249994	0.2627078	0.2573680	0.2647860	0.414471
2	rowSums	0.406690	0.449248	0.4665504	0.4651525	0.4743145	0.603410
3	apply+sum	2.053905	2.266421	2.5475558	2.3270635	2.3745725	20.193188

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	1.791507	1.797035	1.775929	1.807344	1.791313	1.455856
3	apply+sum	9.047641	9.065904	9.697297	9.041775	8.967893	48.720388

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
2	rowCounts	227.010	249.9940	262.7078	257.368	264.786	414.471
1	colCounts	307.298	337.6245	350.5225	347.547	355.038	528.144

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.00000	1.000000	1.000000	1.000000	1.00000
1	colCounts	1.353676	1.35053	1.334267	1.350389	1.340849	1.27426

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
2	colSums	0.264082	0.2834590	0.3026913	0.2948455	0.3111115	0.445120
1	colCounts	0.294222	0.3198440	0.3525906	0.3305465	0.3760595	0.527161
3	apply+sum	0.865563	0.9335955	1.0999853	0.9697700	1.1156315	8.721477

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.114131	1.128361	1.164852	1.121084	1.208761	1.184312
3	apply+sum	3.277630	3.293582	3.634017	3.289079	3.585954	19.593541

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	0.230868	0.2524785	0.2713328	0.2598240	0.289090	0.404084
2	rowSums	0.507178	0.5500880	0.5884659	0.5672945	0.600566	0.858183
3	apply+sum	0.868879	0.9416785	1.1061136	0.9848270	1.086796	8.908098

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	2.196831	2.178752	2.168797	2.183380	2.077436	2.123774
3	apply+sum	3.763532	3.729737	4.076594	3.790362	3.759369	22.045164

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
2	rowCounts	230.868	252.4785	271.3328	259.8240	289.0900	404.084
1	colCounts	294.222	319.8440	352.5906	330.5465	376.0595	527.161

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.274417	1.266817	1.299476	1.272194	1.300839	1.304583

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	0.006344	0.0071695	0.0081452	0.0081490	0.0087110	0.019743
1	colCounts	0.007943	0.0087810	0.0101203	0.0095020	0.0108475	0.039544
3	apply+sum	0.041412	0.0426590	0.0445136	0.0436835	0.0447230	0.109605

	expr	min	lq	mean	median	uq	max
2	colSums	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCounts	1.252049	1.224772	1.242481	1.166033	1.245265	2.002938
3	apply+sum	6.527743	5.950066	5.464980	5.360596	5.134083	5.551588

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
2	rowSums	0.007392	0.007844	0.0087515	0.008557	0.0092035	0.020365
1	rowCounts	0.006947	0.007600	0.0089012	0.009197	0.0096325	0.032405
3	apply+sum	0.041534	0.042505	0.0441844	0.043150	0.0438720	0.109962

	expr	min	lq	mean	median	uq	max
2	rowSums	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	rowCounts	0.9397998	0.9688934	1.017106	1.074793	1.046613	1.591210
3	apply+sum	5.6187771	5.4187914	5.048778	5.042655	4.766882	5.399558

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	6.947	7.600	8.90121	9.197	9.6325	32.405
1	colCounts	7.943	8.781	10.12032	9.502	10.8475	39.544

	expr	min	lq	mean	median	uq	max
2	rowCounts	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
1	colCounts	1.143371	1.155395	1.13696	1.033163	1.126135	1.220306

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	0.029451	0.0327940	0.0364990	0.0345375	0.0390410	0.084870
2	colSums	0.034015	0.0366715	0.0405585	0.0387515	0.0427565	0.062180
3	apply+sum	0.259060	0.2786255	0.3076613	0.2882320	0.3260625	0.462047

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colSums	1.154969	1.118238	1.111224	1.122012	1.095169	0.7326499
3	apply+sum	8.796306	8.496234	8.429312	8.345480	8.351797	5.4441734

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	0.034162	0.0376705	0.0413610	0.0395395	0.0449590	0.074276
2	rowSums	0.061404	0.0658195	0.0718368	0.0688490	0.0761715	0.098664
3	apply+sum	0.258734	0.2785950	0.3077394	0.2928375	0.3306725	0.501462

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	1.797436	1.747243	1.736824	1.741271	1.694244	1.328343
3	apply+sum	7.573737	7.395575	7.440327	7.406201	7.354979	6.751333

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
1	colCounts	29.451	32.7940	36.49898	34.5375	39.041	84.870
2	rowCounts	34.162	37.6705	41.36100	39.5395	44.959	74.276

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.00000	1.000000	1.000000	1.0000000
2	rowCounts	1.159961	1.148701	1.13321	1.144828	1.151584	0.8751738

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	0.032257	0.0350105	0.0383101	0.0381240	0.0405645	0.069719
2	colSums	0.036943	0.0411305	0.0443320	0.0426715	0.0471730	0.080673
3	apply+sum	0.134531	0.1494320	0.1595826	0.1582715	0.1703885	0.255708

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.145271	1.174805	1.157189	1.119282	1.162913	1.157116
3	apply+sum	4.170599	4.268205	4.165552	4.151492	4.200434	3.667695

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	0.035295	0.0384135	0.0423436	0.0417395	0.0461890	0.073065
3	apply+sum	0.126264	0.1350770	0.1465840	0.1405220	0.1551805	0.228310
2	rowSums	0.183670	0.1957190	0.2099086	0.1985195	0.2278255	0.277979

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	apply+sum	3.577391	3.516394	3.461779	3.366643	3.359685	3.124752
2	rowSums	5.203853	5.095058	4.957275	4.756154	4.932462	3.804544

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	32.257	35.0105	38.31008	38.1240	40.5645	69.719
2	rowCounts	35.295	38.4135	42.34355	41.7395	46.1890	73.065

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	1.094181	1.097199	1.105285	1.094835	1.138656	1.047993

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	0.026414	0.0304460	0.0347676	0.0328335	0.036252	0.079650
2	colSums	0.028785	0.0321795	0.0374942	0.0349530	0.039920	0.081596
3	apply+sum	1.331657	1.4318335	1.5451198	1.4898125	1.598719	2.168075

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.089763	1.056937	1.078424	1.064553	1.101181	1.024432
3	apply+sum	50.414818	47.028624	44.441371	45.374770	44.100160	27.220025

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	0.029032	0.0331615	0.0380870	0.035666	0.039444	0.091821
2	rowSums	0.043969	0.0476690	0.0522297	0.049749	0.054849	0.090145
3	apply+sum	1.329039	1.4252550	1.5246144	1.494509	1.580462	2.077634

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
2	rowSums	1.514501	1.43748	1.371327	1.394858	1.390554	0.9817471
3	apply+sum	45.778417	42.97921	40.029795	41.902919	40.068502	22.6270025

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
1	colCounts	26.414	30.4460	34.76760	32.8335	36.252	79.650
2	rowCounts	29.032	33.1615	38.08699	35.6660	39.444	91.821

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowCounts	1.099114	1.089191	1.095474	1.086269	1.08805	1.152806

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	0.204109	0.2176435	0.2348922	0.2308800	0.2431070	0.351074
2	colSums	0.236322	0.2598745	0.2830678	0.2720045	0.2954255	0.446940
3	apply+sum	2.061638	2.2621880	2.5896254	2.3428130	2.5183635	18.981519

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.157822	1.194038	1.205097	1.178121	1.215208	1.273065
3	apply+sum	10.100672	10.394007	11.024739	10.147319	10.359074	54.067003

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	0.244753	0.2655530	0.2798852	0.271549	0.2849060	0.444187
2	rowSums	0.379399	0.4117765	0.4325325	0.425415	0.4430005	0.586766
3	apply+sum	2.050156	2.2776410	2.5832776	2.336193	2.4756675	18.869115

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	1.550130	1.550638	1.545392	1.566623	1.554901	1.320989
3	apply+sum	8.376428	8.576973	9.229775	8.603208	8.689419	42.480115

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	204.109	217.6435	234.8922	230.880	243.107	351.074
2	rowCounts	244.753	265.5530	279.8852	271.549	284.906	444.187

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCounts	1.199129	1.220128	1.191547	1.176148	1.171937	1.265223

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	0.188446	0.2039120	0.2266170	0.2138475	0.2439680	0.363926
2	colSums	0.239284	0.2524595	0.2764089	0.2650605	0.2946225	0.394276
3	apply+sum	0.869810	0.9181520	1.0874815	0.9709860	1.0909525	7.299217

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colSums	1.269775	1.238081	1.219718	1.239484	1.207628	1.083396
3	apply+sum	4.615699	4.502687	4.798764	4.540553	4.471703	20.056871

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	0.245975	0.2657100	0.2879802	0.2738550	0.294662	0.552648
2	rowSums	0.478853	0.5219055	0.6137815	0.5359280	0.568132	6.887381
3	apply+sum	0.895298	0.9454045	1.0431311	0.9930245	1.098558	2.157093

	expr	min	lq	mean	median	uq	max
1	rowCounts	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowSums	1.946755	1.964192	2.131332	1.956977	1.928080	12.462510
3	apply+sum	3.639793	3.558031	3.622232	3.626096	3.728199	3.903195

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
1	colCounts	188.446	203.912	226.6170	213.8475	243.968	363.926
2	rowCounts	245.975	265.710	287.9802	273.8550	294.662	552.648

	expr	min	lq	mean	median	uq	max
1	colCounts	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowCounts	1.305281	1.303062	1.270779	1.280609	1.20779	1.518573

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_0.60.1     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] rappdirs_0.3.3          startup_0.15.0          labeling_0.4.2         
[67] bitops_1.0-7            base64enc_0.1-3         boot_1.3-28            
[70] gtable_0.3.0            DBI_1.1.1               markdown_1.1           
[73] R6_2.5.1                lpSolveAPI_5.5.2.0-17.7 rle_0.9.2              
[76] dplyr_1.0.7             fastmap_1.1.0           bit_4.0.4              
[79] utf8_1.2.2              parallel_4.1.1          Rcpp_1.0.7             
[82] vctrs_0.3.8             png_0.1-7               DEoptimR_1.0-9         
[85] tidyselect_1.1.1        xfun_0.25               coda_0.19-4            

Total processing time was 38.2 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Henrik Bengtsson. Last updated on 2021-08-25 18:52:18 (+0200 UTC). Powered by RSP.