matrixStats.benchmarks

---

colAnyMissings() and rowAnyMissings() benchmarks on subsetted computation

This report benchmark the performance of colAnyMissings() and rowAnyMissings() on subsetted computation.

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"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5168238 276.1    8529671 455.6  8529671 455.6
Vcells 9405490  71.8   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160373 275.6    8529671 455.6  8529671 455.6
Vcells 9379366  71.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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
1	colAnyMissings_X_S	0.004764	0.0048960	0.0081695	0.0049975	0.0051465	0.300122
2	colAnyMissings(X, rows, cols)	0.005137	0.0052965	0.0056056	0.0053910	0.0055315	0.020641
3	colAnyMissings(X[rows, cols])	0.005829	0.0061720	0.0064292	0.0062995	0.0064405	0.011839

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.078296	1.081802	0.6861648	1.078739	1.074808	0.0687754
3	colAnyMissings(X[rows, cols])	1.223552	1.260621	0.7869713	1.260530	1.251433	0.0394473

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.004935	0.0050450	0.0051830	0.0051470	0.0052395	0.007354
2	rowAnyMissings(X, cols, rows)	0.005243	0.0054750	0.0085878	0.0055525	0.0057245	0.299638
3	rowAnyMissings(X[cols, rows])	0.005927	0.0062455	0.0064763	0.0064115	0.0065780	0.009756

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.062411	1.085233	1.656895	1.078784	1.092566	40.744901
3	rowAnyMissings(X[cols, rows])	1.201013	1.237958	1.249517	1.245677	1.255463	1.326625

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+10x10 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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
1	colAnyMissings_X_S	4.764	4.896	8.16951	4.9975	5.1465	300.122
2	rowAnyMissings_X_S	4.935	5.045	5.18305	5.1470	5.2395	7.354

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowAnyMissings_X_S	1.035894	1.030433	0.6344383	1.029915	1.018071	0.0245034

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

100x100 matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5159500 275.6    8529671 455.6  8529671 455.6
Vcells 9214159  70.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5159476 275.6    8529671 455.6  8529671 455.6
Vcells 9219212  70.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.013782	0.0141225	0.0144279	0.0143485	0.0145980	0.020174
2	colAnyMissings(X, rows, cols)	0.017009	0.0176770	0.0180100	0.0179010	0.0181995	0.022137
3	colAnyMissings(X[rows, cols])	0.025211	0.0257895	0.0271811	0.0260085	0.0263985	0.062706

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.234146	1.251691	1.248279	1.247587	1.246712	1.097303
3	colAnyMissings(X[rows, cols])	1.829270	1.826129	1.883927	1.812628	1.808364	3.108258

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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
2	rowAnyMissings(X, cols, rows)	0.013759	0.0147005	0.0160306	0.0149900	0.0155905	0.052725
1	rowAnyMissings_X_S	0.013815	0.0150885	0.0160177	0.0153965	0.0159835	0.035255
3	rowAnyMissings(X[cols, rows])	0.024877	0.0264115	0.0274820	0.0268960	0.0280825	0.042618

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings(X, cols, rows)	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowAnyMissings_X_S	1.004070	1.026394	0.999194	1.027118	1.025208	0.6686581
3	rowAnyMissings(X[cols, rows])	1.808053	1.796640	1.714341	1.794263	1.801257	0.8083073

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+100x100 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	13.782	14.1225	14.42787	14.3485	14.5980	20.174
2	rowAnyMissings_X_S	13.815	15.0885	16.01771	15.3965	15.9835	35.255

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowAnyMissings_X_S	1.002394	1.068402	1.110192	1.073039	1.09491	1.747546

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

1000x10 matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160242 275.6    8529671 455.6  8529671 455.6
Vcells 9218209  70.4   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160218 275.6    8529671 455.6  8529671 455.6
Vcells 9223262  70.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.012746	0.0133290	0.0137894	0.0135525	0.0136995	0.027639
2	colAnyMissings(X, rows, cols)	0.017872	0.0187040	0.0191630	0.0189465	0.0192710	0.026558
3	colAnyMissings(X[rows, cols])	0.024450	0.0255475	0.0264699	0.0259050	0.0261865	0.062811

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.402165	1.403256	1.389694	1.398008	1.406694	0.9608886
3	colAnyMissings(X[rows, cols])	1.918249	1.916685	1.919589	1.911455	1.911493	2.2725497

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.015884	0.0166435	0.0170975	0.0168815	0.0171135	0.029345
2	rowAnyMissings(X, cols, rows)	0.016958	0.0177320	0.0184489	0.0181270	0.0186145	0.024209
3	rowAnyMissings(X[cols, rows])	0.029481	0.0307810	0.0316317	0.0312070	0.0314415	0.070962

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowAnyMissings(X, cols, rows)	1.067615	1.065401	1.079041	1.073779	1.087708	0.8249787
3	rowAnyMissings(X[cols, rows])	1.856019	1.849431	1.850078	1.848592	1.837234	2.4181973

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+1000x10 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	12.746	13.3290	13.78937	13.5525	13.6995	27.639
2	rowAnyMissings_X_S	15.884	16.6435	17.09748	16.8815	17.1135	29.345

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings_X_S	1.246195	1.248668	1.239903	1.245637	1.249206	1.061724

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

10x1000 matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160447 275.6    8529671 455.6  8529671 455.6
Vcells 9219140  70.4   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160423 275.6    8529671 455.6  8529671 455.6
Vcells 9224193  70.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.015783	0.0167735	0.0174614	0.0169845	0.0173080	0.047039
2	colAnyMissings(X, rows, cols)	0.021038	0.0226015	0.0229454	0.0228350	0.0230775	0.034074
3	colAnyMissings(X[rows, cols])	0.029427	0.0308190	0.0315096	0.0310815	0.0316075	0.046379

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.332953	1.347453	1.314064	1.344461	1.333343	0.7243776
3	colAnyMissings(X[rows, cols])	1.864474	1.837363	1.804527	1.829992	1.826179	0.9859691

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.016019	0.0163735	0.0167852	0.0165925	0.0168880	0.022359
2	rowAnyMissings(X, cols, rows)	0.018313	0.0188270	0.0195425	0.0191855	0.0194705	0.054754
3	rowAnyMissings(X[cols, rows])	0.028022	0.0284220	0.0287956	0.0286640	0.0290390	0.034448

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.143205	1.149846	1.164274	1.156275	1.152919	2.448857
3	rowAnyMissings(X[cols, rows])	1.749298	1.735854	1.715537	1.727528	1.719505	1.540677

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+10x1000 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	rowAnyMissings_X_S	16.019	16.3735	16.78516	16.5925	16.888	22.359
1	colAnyMissings_X_S	15.783	16.7735	17.46142	16.9845	17.308	47.039

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings_X_S	1.0000000	1.00000	1.000000	1.000000	1.00000	1.000000
1	colAnyMissings_X_S	0.9852675	1.02443	1.040289	1.023625	1.02487	2.103806

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

100x1000 matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160658 275.7    8529671 455.6  8529671 455.6
Vcells 9241833  70.6   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160634 275.7    8529671 455.6  8529671 455.6
Vcells 9291886  70.9   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.068732	0.0718555	0.0796662	0.0765295	0.085540	0.123166
2	colAnyMissings(X, rows, cols)	0.087344	0.0909465	0.1007715	0.0969410	0.109387	0.171444
3	colAnyMissings(X[rows, cols])	0.149315	0.1598135	0.1732749	0.1712850	0.186491	0.212085

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.270791	1.265686	1.264922	1.266714	1.278782	1.391975
3	colAnyMissings(X[rows, cols])	2.172423	2.224096	2.175013	2.238157	2.180161	1.721944

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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
2	rowAnyMissings(X, cols, rows)	0.069810	0.0726365	0.0796728	0.0748425	0.0847790	0.142066
1	rowAnyMissings_X_S	0.069450	0.0741070	0.0810793	0.0801935	0.0865870	0.124561
3	rowAnyMissings(X[cols, rows])	0.146539	0.1560045	0.1687410	0.1629055	0.1811055	0.201860

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings(X, cols, rows)	1.0000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowAnyMissings_X_S	0.9948431	1.020245	1.017654	1.071497	1.021326	0.8767826
3	rowAnyMissings(X[cols, rows])	2.0991119	2.147743	2.117924	2.176644	2.136207	1.4208889

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+100x1000 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	68.732	71.8555	79.66615	76.5295	85.540	123.166
2	rowAnyMissings_X_S	69.450	74.1070	81.07934	80.1935	86.587	124.561

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowAnyMissings_X_S	1.010446	1.031334	1.017739	1.047877	1.01224	1.011326

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

1000x100 matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160868 275.7    8529671 455.6  8529671 455.6
Vcells 9242640  70.6   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5160844 275.7    8529671 455.6  8529671 455.6
Vcells 9292693  70.9   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.067708	0.0701475	0.0768881	0.0726405	0.0807180	0.121301
2	colAnyMissings(X, rows, cols)	0.086963	0.0897345	0.1004927	0.0991790	0.1074625	0.149402
3	colAnyMissings(X[rows, cols])	0.145024	0.1504565	0.1688668	0.1608425	0.1866255	0.258294

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.284383	1.279226	1.307000	1.365340	1.331333	1.231663
3	colAnyMissings(X[rows, cols])	2.141904	2.144859	2.196267	2.214226	2.312068	2.129364

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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
2	rowAnyMissings(X, cols, rows)	0.064863	0.0716550	0.0772394	0.0743695	0.0788165	0.176123
1	rowAnyMissings_X_S	0.069201	0.0755755	0.0802294	0.0787945	0.0835245	0.108392
3	rowAnyMissings(X[cols, rows])	0.143929	0.1572250	0.1682894	0.1633200	0.1739550	0.227019

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings(X, cols, rows)	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowAnyMissings_X_S	1.066879	1.054714	1.038710	1.059500	1.059734	0.6154335
3	rowAnyMissings(X[cols, rows])	2.218969	2.194194	2.178801	2.196062	2.207089	1.2889799

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on integer+1000x100 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	67.708	70.1475	76.88810	72.6405	80.7180	121.301
2	rowAnyMissings_X_S	69.201	75.5755	80.22942	78.7945	83.5245	108.392

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.00000	1.000000	1.000000	1.000000	1.0000000
2	rowAnyMissings_X_S	1.022051	1.07738	1.043457	1.084719	1.034769	0.8935788

Figure: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161086 275.7    8529671 455.6  8529671 455.6
Vcells 9333779  71.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161053 275.7    8529671 455.6  8529671 455.6
Vcells 9333917  71.3   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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
1	colAnyMissings_X_S	0.004763	0.0049160	0.0053280	0.0050130	0.0050915	0.031773
2	colAnyMissings(X, rows, cols)	0.005136	0.0053475	0.0054884	0.0054315	0.0055435	0.008306
3	colAnyMissings(X[rows, cols])	0.005850	0.0062080	0.0063675	0.0062870	0.0064075	0.011304

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.078312	1.087775	1.030103	1.083483	1.088775	0.2614169
3	colAnyMissings(X[rows, cols])	1.228217	1.262815	1.195118	1.254139	1.258470	0.3557738

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.004928	0.005112	0.0052579	0.0052035	0.0053700	0.007703
2	rowAnyMissings(X, cols, rows)	0.005289	0.005462	0.0059575	0.0055770	0.0056985	0.034002
3	rowAnyMissings(X[cols, rows])	0.006113	0.006315	0.0065227	0.0064595	0.0065995	0.009716

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.073255	1.068466	1.133066	1.071779	1.061173	4.414124
3	rowAnyMissings(X[cols, rows])	1.240463	1.235329	1.240547	1.241376	1.228957	1.261327

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+10x10 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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
1	colAnyMissings_X_S	4.763	4.916	5.32797	5.0130	5.0915	31.773
2	rowAnyMissings_X_S	4.928	5.112	5.25790	5.2035	5.3700	7.703

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.00000	1.0000000	1.000000	1.000000	1.0000000
2	rowAnyMissings_X_S	1.034642	1.03987	0.9868486	1.038001	1.054699	0.2424385

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

100x100 matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161285 275.7    8529671 455.6  8529671 455.6
Vcells 9339765  71.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161261 275.7    8529671 455.6  8529671 455.6
Vcells 9349818  71.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.014080	0.0147540	0.0153998	0.0151635	0.0155990	0.030410
2	colAnyMissings(X, rows, cols)	0.014756	0.0155700	0.0162705	0.0161000	0.0166425	0.022333
3	colAnyMissings(X[rows, cols])	0.029500	0.0308305	0.0319412	0.0313230	0.0323670	0.069307

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.048011	1.055307	1.056539	1.061760	1.066895	0.7343966
3	colAnyMissings(X[rows, cols])	2.095171	2.089637	2.074134	2.065684	2.074941	2.2790858

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.017863	0.0185890	0.0194330	0.0193800	0.0200885	0.025795
2	rowAnyMissings(X, cols, rows)	0.018339	0.0191705	0.0204793	0.0199875	0.0206215	0.056908
3	rowAnyMissings(X[cols, rows])	0.033330	0.0341825	0.0358083	0.0354035	0.0369535	0.049152

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.026647	1.031282	1.053840	1.031347	1.026533	2.206164
3	rowAnyMissings(X[cols, rows])	1.865868	1.838856	1.842654	1.826806	1.839535	1.905486

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+100x100 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	14.080	14.754	15.39977	15.1635	15.5990	30.410
2	rowAnyMissings_X_S	17.863	18.589	19.43298	19.3800	20.0885	25.795

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowAnyMissings_X_S	1.268679	1.259929	1.261901	1.278069	1.287807	0.8482407

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

1000x10 matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161484 275.7    8529671 455.6  8529671 455.6
Vcells 9341215  71.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161460 275.7    8529671 455.6  8529671 455.6
Vcells 9351268  71.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.012883	0.0134635	0.0138016	0.0136990	0.0138585	0.025085
2	colAnyMissings(X, rows, cols)	0.016309	0.0170090	0.0175423	0.0172810	0.0176415	0.031930
3	colAnyMissings(X[rows, cols])	0.029507	0.0306820	0.0313238	0.0310105	0.0312045	0.068240

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.265932	1.263342	1.271037	1.261479	1.272973	1.272872
3	colAnyMissings(X[rows, cols])	2.290383	2.278902	2.269578	2.263705	2.251651	2.720351

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.017414	0.0189950	0.0195016	0.0193190	0.0196165	0.034245
2	rowAnyMissings(X, cols, rows)	0.019997	0.0210155	0.0215119	0.0217030	0.0219970	0.024958
3	rowAnyMissings(X[cols, rows])	0.036654	0.0385975	0.0399305	0.0398395	0.0400455	0.081435

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowAnyMissings(X, cols, rows)	1.148329	1.106370	1.103087	1.123402	1.121352	0.7288071
3	rowAnyMissings(X[cols, rows])	2.104858	2.031982	2.047554	2.062193	2.041419	2.3780114

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+1000x10 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	12.883	13.4635	13.80159	13.699	13.8585	25.085
2	rowAnyMissings_X_S	17.414	18.9950	19.50156	19.319	19.6165	34.245

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings_X_S	1.351704	1.410852	1.412994	1.410249	1.415485	1.365158

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

10x1000 matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161689 275.7    8529671 455.6  8529671 455.6
Vcells 9341351  71.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161665 275.7    8529671 455.6  8529671 455.6
Vcells 9351404  71.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.016341	0.0178340	0.0186672	0.0184625	0.0188980	0.046513
2	colAnyMissings(X, rows, cols)	0.019917	0.0210530	0.0221198	0.0220160	0.0224810	0.039709
3	colAnyMissings(X[rows, cols])	0.035556	0.0374405	0.0385000	0.0387920	0.0391845	0.053957

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.218836	1.180498	1.184957	1.192471	1.189597	0.8537183
3	colAnyMissings(X[rows, cols])	2.175877	2.099389	2.062445	2.101124	2.073473	1.1600413

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.019145	0.0198565	0.0206390	0.020370	0.021120	0.028078
2	rowAnyMissings(X, cols, rows)	0.021345	0.0226535	0.0239933	0.023574	0.024279	0.061634
3	rowAnyMissings(X[cols, rows])	0.034874	0.0363200	0.0374831	0.037167	0.038499	0.050534

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.114912	1.140861	1.162522	1.157290	1.149574	2.195099
3	rowAnyMissings(X[cols, rows])	1.821572	1.829124	1.816129	1.824595	1.822869	1.799772

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+10x1000 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	16.341	17.8340	18.66715	18.4625	18.898	46.513
2	rowAnyMissings_X_S	19.145	19.8565	20.63900	20.3700	21.120	28.078

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowAnyMissings_X_S	1.171593	1.113407	1.105632	1.103317	1.117579	0.6036592

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

100x1000 matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161900 275.7    8529671 455.6  8529671 455.6
Vcells 9386864  71.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5161876 275.7    8529671 455.6  8529671 455.6
Vcells 9486917  72.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.073336	0.078950	0.0860482	0.0819950	0.0909950	0.136407
2	colAnyMissings(X, rows, cols)	0.081648	0.087093	0.0957248	0.0904085	0.1028235	0.204545
3	colAnyMissings(X[rows, cols])	0.187117	0.199815	0.2178751	0.2079400	0.2298140	0.263287

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.113341	1.103141	1.112455	1.102610	1.129991	1.499520
3	colAnyMissings(X[rows, cols])	2.551503	2.530906	2.532011	2.536008	2.525567	1.930157

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.089237	0.100313	0.1108312	0.1079435	0.1195555	0.166717
2	rowAnyMissings(X, cols, rows)	0.094051	0.102488	0.1126039	0.1088550	0.1209705	0.226227
3	rowAnyMissings(X[cols, rows])	0.190255	0.210242	0.2330036	0.2287820	0.2520205	0.318517

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.053946	1.021682	1.015994	1.008444	1.011836	1.356952
3	rowAnyMissings(X[cols, rows])	2.132019	2.095860	2.102328	2.119461	2.107979	1.910525

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+100x1000 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	73.336	78.950	86.04824	81.9950	90.9950	136.407
2	rowAnyMissings_X_S	89.237	100.313	110.83122	107.9435	119.5555	166.717

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings_X_S	1.216824	1.270589	1.288013	1.316464	1.313869	1.222203

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

1000x100 matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5162110 275.7    8529671 455.6  8529671 455.6
Vcells 9387005  71.7   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colAnyMissings_X_S = colAnyMissings(X_S), `colAnyMissings(X, rows, cols)` = colAnyMissings(X, 
+     rows = rows, cols = cols), `colAnyMissings(X[rows, cols])` = colAnyMissings(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5162086 275.7    8529671 455.6  8529671 455.6
Vcells 9487058  72.4   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowAnyMissings_X_S = rowAnyMissings(X_S), `rowAnyMissings(X, cols, rows)` = rowAnyMissings(X, 
+     rows = cols, cols = rows), `rowAnyMissings(X[cols, rows])` = rowAnyMissings(X[cols, rows]), unit = "ms")

Table: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() 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	colAnyMissings_X_S	0.068125	0.070902	0.0776889	0.073456	0.0829200	0.127949
2	colAnyMissings(X, rows, cols)	0.077436	0.081009	0.0905127	0.087444	0.0965605	0.139821
3	colAnyMissings(X[rows, cols])	0.181127	0.190029	0.2118699	0.203866	0.2336380	0.340141

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	1.136675	1.142549	1.165067	1.190427	1.164502	1.092787
3	colAnyMissings(X[rows, cols])	2.658745	2.680164	2.727158	2.775349	2.817632	2.658411

Table: Benchmarking of rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() 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	rowAnyMissings_X_S	0.093433	0.1024075	0.1131587	0.1093515	0.1197825	0.209474
2	rowAnyMissings(X, cols, rows)	0.093000	0.1052835	0.1163487	0.1120100	0.1245360	0.277027
3	rowAnyMissings(X[cols, rows])	0.195649	0.2170510	0.2428237	0.2353920	0.2577630	0.407847

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	0.9953657	1.028084	1.028190	1.024312	1.039684	1.322489
3	rowAnyMissings(X[cols, rows])	2.0940032	2.119483	2.145868	2.152618	2.151925	1.947005

Figure: Benchmarking of colAnyMissings_X_S(), colAnyMissings(X, rows, cols)() and colAnyMissings(X[rows, cols])() on double+1000x100 data as well as rowAnyMissings_X_S(), rowAnyMissings(X, cols, rows)() and rowAnyMissings(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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	colAnyMissings_X_S	68.125	70.9020	77.68889	73.4560	82.9200	127.949
2	rowAnyMissings_X_S	93.433	102.4075	113.15871	109.3515	119.7825	209.474

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings_X_S	1.371494	1.444353	1.456562	1.488667	1.444555	1.637168

Figure: Benchmarking of colAnyMissings_X_S() and rowAnyMissings_X_S() 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 22.17 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Dongcan Jiang. Last updated on 2021-08-25 18:50:12 (+0200 UTC). Powered by RSP.