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 5179685 276.7    7916910 422.9  7916910 422.9
Vcells 9438286  72.1   33191153 253.3 53339345 407.0
> 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 5167805 276.0    7916910 422.9  7916910 422.9
Vcells 9398484  71.8   33191153 253.3 53339345 407.0
> 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	0.004148	0.0043350	0.0080678	0.0044310	0.0045380	0.328684
2	colAnyMissings(X, rows, cols)	0.004449	0.0046215	0.0049160	0.0047235	0.0048575	0.018754
3	colAnyMissings(X[rows, cols])	0.005104	0.0053415	0.0055997	0.0054735	0.0056145	0.012662

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.00000	1.0000000	1.000000	1.000000	1.0000000
2	colAnyMissings(X, rows, cols)	1.072565	1.06609	0.6093399	1.066012	1.070405	0.0570578
3	colAnyMissings(X[rows, cols])	1.230473	1.23218	0.6940856	1.235274	1.237219	0.0385233

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	0.004460	0.0045710	0.0047036	0.0046575	0.0047450	0.007701
2	rowAnyMissings(X, cols, rows)	0.004767	0.0049695	0.0083557	0.0050895	0.0052170	0.327541
3	rowAnyMissings(X[cols, rows])	0.005499	0.0056820	0.0058936	0.0058300	0.0059525	0.008993

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
2	rowAnyMissings(X, cols, rows)	1.068834	1.087180	1.776434	1.092754	1.099473	42.53227
3	rowAnyMissings(X[cols, rows])	1.232960	1.243054	1.252996	1.251745	1.254478	1.16777

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	4.148	4.335	8.06778	4.4310	4.538	328.684
2	rowAnyMissings_X_S	4.460	4.571	4.70363	4.6575	4.745	7.701

	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	1.075217	1.054441	0.5830142	1.051117	1.045615	0.0234298

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 5166984 276.0    7916910 422.9  7916910 422.9
Vcells 9233416  70.5   33191153 253.3 53339345 407.0
> 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 5166978 276.0    7916910 422.9  7916910 422.9
Vcells 9238499  70.5   33191153 253.3 53339345 407.0
> 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	0.007100	0.0077060	0.0079425	0.0079700	0.0080495	0.011545
2	colAnyMissings(X, rows, cols)	0.013799	0.0144120	0.0146853	0.0145945	0.0147980	0.018575
3	colAnyMissings(X[rows, cols])	0.018313	0.0192085	0.0199999	0.0194345	0.0197345	0.048600

	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)	1.943521	1.870231	1.848942	1.831179	1.838375	1.608922
3	colAnyMissings(X[rows, cols])	2.579296	2.492668	2.518074	2.438457	2.451643	4.209615

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
1	rowAnyMissings_X_S	0.010208	0.010779	0.0112918	0.0113265	0.011609	0.017024
2	rowAnyMissings(X, cols, rows)	0.011866	0.012345	0.0141362	0.0130385	0.013595	0.067164
3	rowAnyMissings(X[cols, rows])	0.021455	0.022036	0.0233730	0.0231360	0.023595	0.034190

	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)	1.162422	1.145282	1.251892	1.151150	1.171074	3.945254
3	rowAnyMissings(X[cols, rows])	2.101783	2.044345	2.069908	2.042643	2.032475	2.008341

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	7.100	7.706	7.94254	7.9700	8.0495	11.545
2	rowAnyMissings_X_S	10.208	10.779	11.29183	11.3265	11.6090	17.024

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.00000	1.00000	1.000000	1.000000	1.000000
2	rowAnyMissings_X_S	1.437747	1.39878	1.42169	1.421142	1.442201	1.474578

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 5167726 276.0    7916910 422.9  7916910 422.9
Vcells 9237476  70.5   33191153 253.3 53339345 407.0
> 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 5167720 276.0    7916910 422.9  7916910 422.9
Vcells 9242559  70.6   33191153 253.3 53339345 407.0
> 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	0.006233	0.0065435	0.0067138	0.0066620	0.0068175	0.009651
2	colAnyMissings(X, rows, cols)	0.013542	0.0141010	0.0146592	0.0142715	0.0144655	0.029932
3	colAnyMissings(X[rows, cols])	0.017324	0.0178665	0.0186964	0.0184085	0.0187205	0.046222

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.00000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colAnyMissings(X, rows, cols)	2.17263	2.154963	2.183439	2.142225	2.121819	3.101440
3	colAnyMissings(X[rows, cols])	2.77940	2.730419	2.784765	2.763209	2.745948	4.789348

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	0.008717	0.0090315	0.0092624	0.0092160	0.009361	0.013308
2	rowAnyMissings(X, cols, rows)	0.013408	0.0138710	0.0142628	0.0141980	0.014468	0.023882
3	rowAnyMissings(X[cols, rows])	0.021336	0.0219555	0.0230095	0.0225535	0.022883	0.053611

	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)	1.538144	1.535847	1.539863	1.540582	1.545561	1.794560
3	rowAnyMissings(X[cols, rows])	2.447631	2.430992	2.484182	2.447211	2.444504	4.028479

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	6.233	6.5435	6.71382	6.662	6.8175	9.651
2	rowAnyMissings_X_S	8.717	9.0315	9.26241	9.216	9.3610	13.308

	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	1.398524	1.380225	1.379604	1.383368	1.373084	1.378925

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 5167931 276.0    7916910 422.9  7916910 422.9
Vcells 9238388  70.5   33191153 253.3 53339345 407.0
> 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 5167925 276.0    7916910 422.9  7916910 422.9
Vcells 9243471  70.6   33191153 253.3 53339345 407.0
> 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	0.009493	0.0106625	0.0131547	0.0112745	0.0149760	0.032852
2	colAnyMissings(X, rows, cols)	0.016530	0.0176355	0.0213137	0.0180620	0.0275785	0.039499
3	colAnyMissings(X[rows, cols])	0.022830	0.0242240	0.0283903	0.0249665	0.0278665	0.049946

	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)	1.741283	1.653974	1.620238	1.602022	1.841513	1.202332
3	colAnyMissings(X[rows, cols])	2.404930	2.271888	2.158188	2.214422	1.860744	1.520334

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	0.009684	0.0101370	0.0106166	0.0106480	0.0108780	0.015548
2	rowAnyMissings(X, cols, rows)	0.013278	0.0138390	0.0146774	0.0144085	0.0148805	0.041930
3	rowAnyMissings(X[cols, rows])	0.020517	0.0211985	0.0225933	0.0220840	0.0231025	0.037550

	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)	1.371128	1.365197	1.382499	1.353165	1.367944	2.696810
3	rowAnyMissings(X[cols, rows])	2.118649	2.091201	2.128116	2.074005	2.123782	2.415102

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	9.684	10.1370	10.61658	10.6480	10.878	15.548
1	colAnyMissings_X_S	9.493	10.6625	13.15470	11.2745	14.976	32.852

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings_X_S	1.0000000	1.00000	1.000000	1.000000	1.000000	1.000000
1	colAnyMissings_X_S	0.9802767	1.05184	1.239071	1.058837	1.376724	2.112941

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 5168142 276.1    7916910 422.9  7916910 422.9
Vcells 9261079  70.7   33191153 253.3 53339345 407.0
> 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 5168136 276.1    7916910 422.9  7916910 422.9
Vcells 9311162  71.1   33191153 253.3 53339345 407.0
> 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	0.023285	0.0245495	0.0268781	0.0256835	0.0279530	0.057528
2	colAnyMissings(X, rows, cols)	0.071859	0.0744675	0.0797039	0.0774855	0.0838680	0.151345
3	colAnyMissings(X[rows, cols])	0.105668	0.1101715	0.1181411	0.1176285	0.1263425	0.138540

	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)	3.086064	3.033361	2.965382	3.016937	3.000322	2.630806
3	colAnyMissings(X[rows, cols])	4.538029	4.487729	4.395438	4.579925	4.519819	2.408219

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
1	rowAnyMissings_X_S	0.038863	0.0404100	0.0436557	0.043490	0.0460685	0.073559
2	rowAnyMissings(X, cols, rows)	0.060459	0.0623870	0.0667166	0.063631	0.0707165	0.122293
3	rowAnyMissings(X[cols, rows])	0.117589	0.1219315	0.1313749	0.126885	0.1405105	0.198744

	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)	1.555696	1.543851	1.528246	1.463118	1.535029	1.662516
3	rowAnyMissings(X[cols, rows])	3.025731	3.017360	3.009344	2.917567	3.050034	2.701831

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	23.285	24.5495	26.87811	25.6835	27.9530	57.528
2	rowAnyMissings_X_S	38.863	40.4100	43.65567	43.4900	46.0685	73.559

	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	1.669014	1.646062	1.624209	1.693305	1.64807	1.278664

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 5168352 276.1    7916910 422.9  7916910 422.9
Vcells 9261885  70.7   33191153 253.3 53339345 407.0
> 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 5168346 276.1    7916910 422.9  7916910 422.9
Vcells 9311968  71.1   33191153 253.3 53339345 407.0
> 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	0.022629	0.0238365	0.0264090	0.0253955	0.0279665	0.058075
2	colAnyMissings(X, rows, cols)	0.073222	0.0763880	0.0835409	0.0845805	0.0888955	0.104431
3	colAnyMissings(X[rows, cols])	0.104353	0.1086115	0.1188780	0.1206140	0.1263750	0.187801

	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)	3.235759	3.204665	3.163352	3.330531	3.178642	1.798209
3	colAnyMissings(X[rows, cols])	4.611472	4.556521	4.501424	4.749424	4.518799	3.233767

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
1	rowAnyMissings_X_S	0.048362	0.0503265	0.0550694	0.0540670	0.058532	0.068993
2	rowAnyMissings(X, cols, rows)	0.063893	0.0665080	0.0732756	0.0712745	0.076519	0.170742
3	rowAnyMissings(X[cols, rows])	0.131056	0.1360320	0.1489086	0.1457050	0.162942	0.183731

	expr	min	lq	mean	median	uq	max
1	rowAnyMissings_X_S	1.000000	1.00000	1.000000	1.000000	1.000000	1.000000
2	rowAnyMissings(X, cols, rows)	1.321141	1.32153	1.330604	1.318262	1.307302	2.474773
3	rowAnyMissings(X[cols, rows])	2.709896	2.70299	2.704016	2.694897	2.783811	2.663038

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	22.629	23.8365	26.40899	25.3955	27.9665	58.075
2	rowAnyMissings_X_S	48.362	50.3265	55.06940	54.0670	58.5320	68.993

	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	2.137169	2.111321	2.085252	2.128999	2.092933	1.187998

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 5168570 276.1    7916910 422.9  7916910 422.9
Vcells 9353023  71.4   33191153 253.3 53339345 407.0
> 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 5168555 276.1    7916910 422.9  7916910 422.9
Vcells 9353191  71.4   33191153 253.3 53339345 407.0
> 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	0.004388	0.0044615	0.0048326	0.0045765	0.0046995	0.025490
2	colAnyMissings(X, rows, cols)	0.004705	0.0049285	0.0050622	0.0050105	0.0051240	0.007731
3	colAnyMissings(X[rows, cols])	0.005384	0.0056545	0.0058520	0.0057630	0.0059200	0.011243

	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)	1.072242	1.104673	1.047511	1.094832	1.090329	0.3032954
3	colAnyMissings(X[rows, cols])	1.226983	1.267399	1.210950	1.259259	1.259708	0.4410749

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	0.004439	0.0045850	0.0047184	0.0046745	0.0048005	0.007243
2	rowAnyMissings(X, cols, rows)	0.004812	0.0049895	0.0058534	0.0050725	0.0052055	0.044818
3	rowAnyMissings(X[cols, rows])	0.005497	0.0057685	0.0059573	0.0058535	0.0060335	0.011667

	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)	1.084028	1.088223	1.240552	1.085143	1.084366	6.187767
3	rowAnyMissings(X[cols, rows])	1.238342	1.258124	1.262572	1.252220	1.256848	1.610797

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	4.388	4.4615	4.83257	4.5765	4.6995	25.490
2	rowAnyMissings_X_S	4.439	4.5850	4.71836	4.6745	4.8005	7.243

	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	1.011623	1.027681	0.9763666	1.021414	1.021492	0.2841506

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 5168769 276.1    7916910 422.9  7916910 422.9
Vcells 9359005  71.5   33191153 253.3 53339345 407.0
> 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 5168763 276.1    7916910 422.9  7916910 422.9
Vcells 9369088  71.5   33191153 253.3 53339345 407.0
> 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	0.008346	0.0091665	0.0094162	0.0093740	0.0095235	0.014397
2	colAnyMissings(X, rows, cols)	0.016231	0.0167725	0.0170888	0.0169465	0.0171470	0.021151
3	colAnyMissings(X[rows, cols])	0.025031	0.0256580	0.0262463	0.0258550	0.0260510	0.058321

	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)	1.944764	1.829760	1.814840	1.807819	1.800493	1.469126
3	colAnyMissings(X[rows, cols])	2.999161	2.799105	2.787364	2.758161	2.735444	4.050913

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	0.009052	0.0096590	0.0101291	0.0099565	0.0101575	0.024910
2	rowAnyMissings(X, cols, rows)	0.014295	0.0149705	0.0156935	0.0153040	0.0156770	0.047241
3	rowAnyMissings(X[cols, rows])	0.025012	0.0263020	0.0264998	0.0264845	0.0266350	0.033584

	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)	1.579209	1.549902	1.549346	1.537086	1.543392	1.896467
3	rowAnyMissings(X[cols, rows])	2.763146	2.723056	2.616191	2.660021	2.622200	1.348214

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	8.346	9.1665	9.41617	9.3740	9.5235	14.397
2	rowAnyMissings_X_S	9.052	9.6590	10.12914	9.9565	10.1575	24.910

	expr	min	lq	mean	median	uq	max
1	colAnyMissings_X_S	1.000000	1.000000	1.000000	1.00000	1.000000	1.000000
2	rowAnyMissings_X_S	1.084591	1.053728	1.075718	1.06214	1.066572	1.730222

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 5168968 276.1    7916910 422.9  7916910 422.9
Vcells 9360456  71.5   33191153 253.3 53339345 407.0
> 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 5168962 276.1    7916910 422.9  7916910 422.9
Vcells 9370539  71.5   33191153 253.3 53339345 407.0
> 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	0.007822	0.0082445	0.0086023	0.0084340	0.0086005	0.018190
2	colAnyMissings(X, rows, cols)	0.016428	0.0169920	0.0172895	0.0172190	0.0174325	0.022876
3	colAnyMissings(X[rows, cols])	0.024324	0.0255395	0.0264378	0.0256535	0.0258835	0.089559

	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)	2.100230	2.061010	2.009872	2.041617	2.026917	1.257614
3	colAnyMissings(X[rows, cols])	3.109691	3.097762	3.073333	3.041676	3.009534	4.923529

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	0.009308	0.009641	0.0098729	0.0098490	0.0099945	0.014667
2	rowAnyMissings(X, cols, rows)	0.016732	0.017037	0.0173575	0.0172730	0.0174600	0.023676
3	rowAnyMissings(X[cols, rows])	0.029381	0.029845	0.0306261	0.0299695	0.0301070	0.073280

	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)	1.797594	1.767140	1.758092	1.753782	1.746961	1.614236
3	rowAnyMissings(X[cols, rows])	3.156532	3.095633	3.102039	3.042898	3.012357	4.996250

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	7.822	8.2445	8.60231	8.434	8.6005	18.190
2	rowAnyMissings_X_S	9.308	9.6410	9.87290	9.849	9.9945	14.667

	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	1.189977	1.169386	1.147703	1.167773	1.162084	0.8063222

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 5169173 276.1    7916910 422.9  7916910 422.9
Vcells 9360592  71.5   33191153 253.3 53339345 407.0
> 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 5169167 276.1    7916910 422.9  7916910 422.9
Vcells 9370675  71.5   33191153 253.3 53339345 407.0
> 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	0.010344	0.0110545	0.0117053	0.0113650	0.0116565	0.038396
2	colAnyMissings(X, rows, cols)	0.019378	0.0207905	0.0213334	0.0211430	0.0215190	0.038683
3	colAnyMissings(X[rows, cols])	0.029667	0.0312820	0.0317273	0.0315315	0.0318480	0.046820

	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)	1.873357	1.880727	1.822539	1.860361	1.846094	1.007475
3	colAnyMissings(X[rows, cols])	2.868039	2.829798	2.710499	2.774439	2.732210	1.219398

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	0.010581	0.0109435	0.0114484	0.0111580	0.0115155	0.020884
2	rowAnyMissings(X, cols, rows)	0.016775	0.0172455	0.0184285	0.0175615	0.0179125	0.048828
3	rowAnyMissings(X[cols, rows])	0.027831	0.0281470	0.0289391	0.0284070	0.0286360	0.045561

	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)	1.585389	1.575867	1.609695	1.573893	1.555512	2.338058
3	rowAnyMissings(X[cols, rows])	2.630281	2.572029	2.527784	2.545886	2.486735	2.181622

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
2	rowAnyMissings_X_S	10.581	10.9435	11.44842	11.158	11.5155	20.884
1	colAnyMissings_X_S	10.344	11.0545	11.70533	11.365	11.6565	38.396

	expr	min	lq	mean	median	uq	max
2	rowAnyMissings_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colAnyMissings_X_S	0.9776014	1.010143	1.022441	1.018552	1.012244	1.838537

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 5169375 276.1    7916910 422.9  7916910 422.9
Vcells 9406085  71.8   33191153 253.3 53339345 407.0
> 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 5169378 276.1    7916910 422.9  7916910 422.9
Vcells 9506183  72.6   33191153 253.3 53339345 407.0
> 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	0.034048	0.0351215	0.0440976	0.0387520	0.0448410	0.146412
2	colAnyMissings(X, rows, cols)	0.091230	0.0922330	0.1107537	0.0990285	0.1154230	0.354018
3	colAnyMissings(X[rows, cols])	0.155771	0.1606975	0.2006325	0.1775685	0.1992585	0.608693

	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)	2.679453	2.626112	2.511562	2.555442	2.574050	2.417957
3	colAnyMissings(X[rows, cols])	4.575041	4.575474	4.549743	4.582176	4.443668	4.157398

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	0.039292	0.0422935	0.0483957	0.0473455	0.0514780	0.084118
2	rowAnyMissings(X, cols, rows)	0.078304	0.0811730	0.0932963	0.0903220	0.1004120	0.195714
3	rowAnyMissings(X[cols, rows])	0.155606	0.1643830	0.1861986	0.1803895	0.2011725	0.322185

	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)	1.992874	1.919278	1.927780	1.907721	1.950581	2.326660
3	rowAnyMissings(X[cols, rows])	3.960246	3.886720	3.847421	3.810066	3.907932	3.830155

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	34.048	35.1215	44.09755	38.7520	44.841	146.412
2	rowAnyMissings_X_S	39.292	42.2935	48.39569	47.3455	51.478	84.118

	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	1.154018	1.204205	1.097469	1.221756	1.148012	0.5745294

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 5169594 276.1    7916910 422.9  7916910 422.9
Vcells 9406241  71.8   33191153 253.3 53339345 407.0
> 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 5169588 276.1    7916910 422.9  7916910 422.9
Vcells 9506324  72.6   33191153 253.3 53339345 407.0
> 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	0.032072	0.0333985	0.0364890	0.0344315	0.0385225	0.078187
2	colAnyMissings(X, rows, cols)	0.087265	0.0906555	0.0989272	0.0970015	0.1054295	0.142899
3	colAnyMissings(X[rows, cols])	0.149542	0.1552250	0.1703300	0.1632410	0.1845645	0.301601

	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)	2.720909	2.714358	2.711151	2.817231	2.736829	1.827657
3	colAnyMissings(X[rows, cols])	4.662696	4.647664	4.667980	4.741037	4.791083	3.857432

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	0.038018	0.0399420	0.0437088	0.0419460	0.0469310	0.064317
2	rowAnyMissings(X, cols, rows)	0.081108	0.0841790	0.0925800	0.0873920	0.0974285	0.226235
3	rowAnyMissings(X[cols, rows])	0.156725	0.1634655	0.1775411	0.1706385	0.1938320	0.220626

	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)	2.133411	2.107531	2.118107	2.083441	2.075995	3.517499
3	rowAnyMissings(X[cols, rows])	4.122389	4.092572	4.061904	4.068052	4.130148	3.430291

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	32.072	33.3985	36.48901	34.4315	38.5225	78.187
2	rowAnyMissings_X_S	38.018	39.9420	43.70883	41.9460	46.9310	64.317

	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	1.185395	1.195922	1.197863	1.218245	1.218275	0.8226048

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_0.60.0     ggplot2_3.3.5         
[4] knitr_1.33             R.devices_2.17.0       R.utils_2.10.1        
[7] R.oo_1.24.0            R.methodsS3_1.8.1-9001 history_0.0.1-9000    

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

Total processing time was 22.25 secs.

Reproducibility

To reproduce this report, do:

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

---

Copyright Dongcan Jiang. Last updated on 2021-08-25 22:10:35 (+0200 UTC). Powered by RSP.