colRowMads - HenrikBengtsson/matrixStats GitHub Wiki

matrixStats: Benchmark report

colMads() and rowMads() benchmarks

This report benchmark the performance of colMads() and rowMads() against alternative methods.

Alternative methods

  • apply() + mad()
  • colMads2() and rowMads2()

where rowMads2() and colMads2() are:

> rowMads2 <- function(x, const = 1.4826, na.rm = FALSE) {
+     mu <- rowMedians(x, na.rm = na.rm)
+     x <- abs(x - mu)
+     mad <- rowMedians(x, na.rm = FALSE)
+     const * mad
+ }
> colMads2 <- function(x, const = 1.4826, na.rm = FALSE) {
+     mu <- colMedians(x, na.rm = na.rm)
+     x <- abs(x - mu)
+     mad <- colMedians(x, na.rm = FALSE)
+     const * mad
+ }

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 integer matrix

> X <- data[["10x10"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132701 167.4    5709258 305.0  5709258 305.0
Vcells 6024971  46.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131392 167.3    5709258 305.0  5709258 305.0
Vcells 6021261  46.0   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.002657 0.0032760 0.0044315 0.0037815 0.0054525 0.016511
2 colMads2 0.004838 0.0056005 0.0083407 0.0069845 0.0096515 0.074925
3 apply+mad 0.489800 0.4961660 0.5043725 0.4998120 0.5051165 0.794902
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.820851 1.709554 1.882139 1.847018 1.770106 4.537884
3 apply+mad 184.343244 151.454823 113.815297 132.172947 92.639432 48.143783

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.002624 0.0032910 0.0044554 0.0042900 0.0055395 0.015129
2 rowMads2 0.004901 0.0057515 0.0083339 0.0067125 0.0093310 0.078219
3 apply+mad 0.489703 0.4945235 0.5008538 0.4989825 0.5020105 0.660943
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.867759 1.747645 1.870525 1.564685 1.684448 5.170137
3 apply+mad 186.624619 150.265421 112.415768 116.312937 90.623793 43.687157

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+10x10 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 2.657 3.276 4.43150 3.7815 5.4525 16.511
2 rowMads 2.624 3.291 4.45537 4.2900 5.5395 15.129
expr min lq mean median uq max
1 colMads 1.00000 1.000000 1.000000 1.00000 1.000000 1.0000000
2 rowMads 0.98758 1.004579 1.005386 1.13447 1.015956 0.9162982

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

100x100 integer matrix

> X <- data[["100x100"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3129993 167.2    5709258 305.0  5709258 305.0
Vcells 5638000  43.1   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3129984 167.2    5709258 305.0  5709258 305.0
Vcells 5643038  43.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.254314 0.257135 0.2647929 0.259611 0.2638700 0.355772
2 colMads2 0.304025 0.306380 0.3141236 0.311071 0.3143015 0.388808
3 apply+mad 5.267928 5.362917 5.7038597 5.439201 5.6298310 13.697193
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.00000 1.000000 1.000000
2 colMads2 1.195471 1.191514 1.186299 1.19822 1.191123 1.092857
3 apply+mad 20.714267 20.856426 21.540835 20.95135 21.335624 38.499918

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.255069 0.2574295 0.2762096 0.2625025 0.2787125 0.409516
2 rowMads2 0.303053 0.3080525 0.3374374 0.3154135 0.3536560 0.535297
3 apply+mad 5.262654 5.3568480 6.0354499 5.4781915 5.8985855 24.659333
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.188122 1.196648 1.221671 1.201564 1.268892 1.307146
3 apply+mad 20.632276 20.808990 21.850977 20.869102 21.163692 60.215799

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+100x100 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 254.314 257.1350 264.7929 259.6110 263.8700 355.772
2 rowMads 255.069 257.4295 276.2096 262.5025 278.7125 409.516
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads 1.002969 1.001145 1.043116 1.011138 1.056249 1.151063

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

1000x10 integer matrix

> X <- data[["1000x10"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3130750 167.3    5709258 305.0  5709258 305.0
Vcells 5641790  43.1   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3130741 167.2    5709258 305.0  5709258 305.0
Vcells 5646828  43.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.184453 0.1861725 0.1888632 0.1873755 0.1886520 0.245527
2 colMads2 0.277145 0.2804365 0.2842387 0.2830035 0.2861835 0.340478
3 apply+mad 0.913187 0.9458865 0.9767638 0.9601225 0.9776690 1.676612
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.502524 1.506326 1.504998 1.510355 1.516992 1.386723
3 apply+mad 4.950784 5.080699 5.171807 5.124056 5.182394 6.828626

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.185629 0.1875140 0.1927187 0.188521 0.189528 0.321136
2 rowMads2 0.288223 0.2909350 0.2999250 0.294080 0.297951 0.478456
3 apply+mad 0.917978 0.9488795 0.9638069 0.961623 0.971357 1.130334
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.552683 1.551538 1.556284 1.559932 1.572069 1.489886
3 apply+mad 4.945229 5.060313 5.001108 5.100880 5.125137 3.519798

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+1000x10 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 184.453 186.1725 188.8632 187.3755 188.652 245.527
2 rowMads 185.629 187.5140 192.7187 188.5210 189.528 321.136
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads 1.006376 1.007206 1.020414 1.006113 1.004644 1.307946

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

10x1000 integer matrix

> X <- data[["10x1000"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3130955 167.3    5709258 305.0  5709258 305.0
Vcells 5642227  43.1   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3130946 167.3    5709258 305.0  5709258 305.0
Vcells 5647265  43.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.255947 0.261325 0.2772449 0.2703955 0.2844750 0.393180
2 colMads2 0.335967 0.341338 0.3611214 0.3587415 0.3705975 0.427025
3 apply+mad 47.194238 48.185322 51.4521724 50.0800570 55.5677890 66.438729
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.00000
2 colMads2 1.312643 1.306182 1.302536 1.326729 1.302742 1.08608
3 apply+mad 184.390667 184.388489 185.583830 185.210394 195.334525 168.97790

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.256452 0.2630255 0.2824898 0.2747510 0.2907425 0.429573
2 rowMads2 0.334503 0.3401210 0.3665499 0.3659265 0.3743685 0.455155
3 apply+mad 46.956961 48.0190850 51.5078001 49.7054005 55.4881420 73.527768
expr min lq mean median uq max
1 rowMads 1.000000 1.00000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.304349 1.29311 1.297568 1.331848 1.287629 1.059552
3 apply+mad 183.102339 182.56437 182.335053 180.910717 190.849779 171.164780

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+10x1000 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() on integer+10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr min lq mean median uq max
1 colMads 255.947 261.3250 277.2449 270.3955 284.4750 393.180
2 rowMads 256.452 263.0255 282.4898 274.7510 290.7425 429.573
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads 1.001973 1.006507 1.018918 1.016108 1.022032 1.092561

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

100x1000 integer matrix

> X <- data[["100x1000"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131168 167.3    5709258 305.0  5709258 305.0
Vcells 5643284  43.1   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131159 167.3    5709258 305.0  5709258 305.0
Vcells 5693322  43.5   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 2.513881 2.520923 2.554454 2.527376 2.543119 3.162973
2 colMads2 2.990316 2.998620 3.050823 3.010641 3.040930 3.719616
3 apply+mad 52.309831 53.302986 57.235065 54.982971 56.280015 70.871133
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.189522 1.189493 1.194315 1.191212 1.195748 1.175987
3 apply+mad 20.808396 21.144230 22.405986 21.754967 22.130307 22.406493

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 2.526592 2.537675 2.593953 2.574283 2.618336 3.079292
2 rowMads2 3.009855 3.030975 3.152807 3.087254 3.134653 4.086106
3 apply+mad 52.563452 53.394471 57.644265 54.967650 59.456479 80.694532
expr min lq mean median uq max
1 rowMads 1.000000 1.00000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.191271 1.19439 1.215445 1.199267 1.197193 1.326963
3 apply+mad 20.804092 21.04070 22.222552 21.352606 22.707734 26.205547

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+100x1000 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 2.513881 2.520923 2.554454 2.527376 2.543119 3.162973
2 rowMads 2.526592 2.537675 2.593953 2.574283 2.618336 3.079292
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.00000 1.000000 1.0000000
2 rowMads 1.005056 1.006645 1.015463 1.01856 1.029576 0.9735436

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

1000x100 integer matrix

> X <- data[["1000x100"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131380 167.3    5709258 305.0  5709258 305.0
Vcells 5643972  43.1   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131371 167.3    5709258 305.0  5709258 305.0
Vcells 5694010  43.5   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 1.964928 1.976131 2.035427 1.988864 2.039416 2.574327
2 colMads2 2.687930 2.720924 2.951148 2.739002 2.779851 10.550021
3 apply+mad 8.982251 9.088460 9.569613 9.228782 9.519751 18.326280
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.367953 1.376894 1.449892 1.377169 1.363063 4.098167
3 apply+mad 4.571288 4.599118 4.701526 4.640227 4.667881 7.118862

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 1.991208 1.998725 2.047768 2.006607 2.022423 3.051313
2 rowMads2 2.758968 2.793220 2.919762 2.830482 2.870141 4.087703
3 apply+mad 8.972996 9.070293 12.046898 9.141897 9.431546 250.482552
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.385575 1.397501 1.425826 1.410581 1.419160 1.339654
3 apply+mad 4.506308 4.538041 5.882940 4.555897 4.663488 82.090088

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on integer+1000x100 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 1.964928 1.976131 2.035427 1.988864 2.039416 2.574327
2 rowMads 1.991208 1.998725 2.047768 2.006607 2.022423 3.051313
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.0000000 1.000000
2 rowMads 1.013375 1.011433 1.006063 1.008921 0.9916677 1.185286

Figure: Benchmarking of colMads() and rowMads() 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 double matrix

> X <- data[["10x10"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131593 167.3    5709258 305.0  5709258 305.0
Vcells 5759841  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131575 167.3    5709258 305.0  5709258 305.0
Vcells 5759964  44.0   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.003801 0.0044480 0.0056274 0.0049490 0.0067115 0.015706
2 colMads2 0.005656 0.0067715 0.0085405 0.0075000 0.0104310 0.018714
3 apply+mad 0.485854 0.4896665 0.4963663 0.4921185 0.4976315 0.652929
expr min lq mean median uq max
1 colMads 1.000000 1.00000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.488029 1.52237 1.517669 1.515458 1.554198 1.191519
3 apply+mad 127.822678 110.08689 88.205260 99.437967 74.146092 41.571947

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.003791 0.0045280 0.0058049 0.0055455 0.0069305 0.017857
2 rowMads2 0.005806 0.0066775 0.0085909 0.0075240 0.0100935 0.020012
3 apply+mad 0.481752 0.4886720 0.5048879 0.4922790 0.4976600 1.082850
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.531522 1.474713 1.479935 1.356776 1.456388 1.120681
3 apply+mad 127.077816 107.922262 86.975712 88.770895 71.807229 60.640085

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+10x10 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 3.801 4.448 5.62740 4.9490 6.7115 15.706
2 rowMads 3.791 4.528 5.80493 5.5455 6.9305 17.857
expr min lq mean median uq max
1 colMads 1.0000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads 0.9973691 1.017986 1.031547 1.120529 1.032631 1.136954

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

100x100 double matrix

> X <- data[["100x100"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131789 167.3    5709258 305.0  5709258 305.0
Vcells 5760747  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131780 167.3    5709258 305.0  5709258 305.0
Vcells 5770785  44.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.349682 0.3518030 0.3636086 0.3545175 0.3583965 0.459405
2 colMads2 0.366045 0.3685025 0.3782240 0.3729490 0.3786860 0.483824
3 apply+mad 5.317847 5.3774485 5.7515005 5.4465390 5.6497875 14.704099
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.00000 1.000000 1.000000
2 colMads2 1.046794 1.047468 1.040195 1.05199 1.056612 1.053154
3 apply+mad 15.207666 15.285397 15.817833 15.36324 15.764070 32.006833

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.349940 0.3518220 0.3597876 0.3541570 0.357123 0.459749
2 rowMads2 0.363962 0.3667865 0.3841819 0.3709645 0.378762 0.497389
3 apply+mad 5.293078 5.3429990 5.6844816 5.4143890 5.605378 14.515281
expr min lq mean median uq max
1 rowMads 1.00000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.04007 1.042534 1.067802 1.047458 1.060593 1.081871
3 apply+mad 15.12567 15.186654 15.799547 15.288104 15.695930 31.572186

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+100x100 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() on double+100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

expr min lq mean median uq max
2 rowMads 349.940 351.822 359.7876 354.1570 357.1230 459.749
1 colMads 349.682 351.803 363.6086 354.5175 358.3965 459.405
expr min lq mean median uq max
2 rowMads 1.0000000 1.000000 1.00000 1.000000 1.000000 1.0000000
1 colMads 0.9992627 0.999946 1.01062 1.001018 1.003566 0.9992518

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

1000x10 double matrix

> X <- data[["1000x10"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132002 167.3    5709258 305.0  5709258 305.0
Vcells 5760891  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3131993 167.3    5709258 305.0  5709258 305.0
Vcells 5770929  44.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.319538 0.320782 0.3246620 0.3223615 0.323855 0.399265
2 colMads2 0.339217 0.342920 0.3488044 0.3464760 0.351911 0.399494
3 apply+mad 1.018295 1.040587 1.0570449 1.0493055 1.061317 1.229027
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.061586 1.069013 1.074362 1.074806 1.086631 1.000574
3 apply+mad 3.186773 3.243909 3.255832 3.255058 3.277136 3.078224

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.321946 0.3238615 0.3290887 0.3253700 0.3276720 0.439519
2 rowMads2 0.351109 0.3544465 0.3616638 0.3564045 0.3604995 0.558510
3 apply+mad 0.976665 0.9961980 1.0128853 1.0049680 1.0122460 1.624371
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.090583 1.094439 1.098986 1.095382 1.100184 1.270730
3 apply+mad 3.033630 3.076000 3.077849 3.088693 3.089205 3.695792

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+1000x10 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 319.538 320.7820 324.6620 322.3615 323.855 399.265
2 rowMads 321.946 323.8615 329.0887 325.3700 327.672 439.519
expr min lq mean median uq max
1 colMads 1.000000 1.0000 1.000000 1.000000 1.000000 1.00000
2 rowMads 1.007536 1.0096 1.013635 1.009333 1.011786 1.10082

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

10x1000 double matrix

> X <- data[["10x1000"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132207 167.3    5709258 305.0  5709258 305.0
Vcells 5761966  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132198 167.3    5709258 305.0  5709258 305.0
Vcells 5772004  44.1   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 0.353791 0.357329 0.3684474 0.3635500 0.3742925 0.428783
2 colMads2 0.394643 0.398898 0.4210527 0.4124335 0.4244375 0.944717
3 apply+mad 46.990760 47.771606 50.8558312 48.7180135 51.6360770 72.002669
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.115469 1.116333 1.142776 1.134462 1.133973 2.203252
3 apply+mad 132.820677 133.690817 138.027385 134.006364 137.956483 167.923329

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 0.353801 0.3583675 0.3719788 0.3648355 0.3774410 0.447620
2 rowMads2 0.394059 0.3991600 0.4224268 0.4144305 0.4283835 0.600624
3 apply+mad 46.881594 47.9615465 51.2745530 49.4159515 54.1775520 68.100929
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.113787 1.113829 1.135621 1.135938 1.134968 1.341817
3 apply+mad 132.508371 133.833415 137.842668 135.447213 143.539128 152.140050

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+10x1000 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 353.791 357.3290 368.4474 363.5500 374.2925 428.783
2 rowMads 353.801 358.3675 371.9788 364.8355 377.4410 447.620
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads 1.000028 1.002906 1.009585 1.003536 1.008412 1.043931

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

100x1000 double matrix

> X <- data[["100x1000"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132420 167.3    5709258 305.0  5709258 305.0
Vcells 5762110  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132411 167.3    5709258 305.0  5709258 305.0
Vcells 5862148  44.8   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 3.551345 3.561309 3.617673 3.580302 3.598743 4.499594
2 colMads2 3.601147 3.615731 3.703038 3.634491 3.700569 4.684895
3 apply+mad 53.368024 54.302672 58.494918 56.169742 58.477301 79.849003
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.014023 1.015281 1.023596 1.015135 1.028295 1.041182
3 apply+mad 15.027553 15.247953 16.169209 15.688549 16.249371 17.745824

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 3.562192 3.583538 3.687924 3.675622 3.722243 4.199921
2 rowMads2 3.634564 3.670988 3.837768 3.750459 3.890684 4.883184
3 apply+mad 53.086682 53.913863 57.987541 55.042718 58.769833 70.578594
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.00000 1.000000 1.000000
2 rowMads2 1.020317 1.024403 1.040631 1.02036 1.045252 1.162685
3 apply+mad 14.902813 15.044870 15.723626 14.97508 15.788820 16.804743

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+100x1000 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 3.551345 3.561309 3.617673 3.580302 3.598743 4.499594
2 rowMads 3.562192 3.583538 3.687924 3.675622 3.722243 4.199921
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.0000
2 rowMads 1.003054 1.006242 1.019419 1.026623 1.034318 0.9334

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

1000x100 double matrix

> X <- data[["1000x100"]]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132632 167.4    5709258 305.0  5709258 305.0
Vcells 5763412  44.0   22267496 169.9 56666022 432.4
> colStats <- microbenchmark(colMads = colMads(X, na.rm = FALSE), colMads2 = colMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 2L, FUN = mad, na.rm = FALSE), unit = "ms")
> X <- t(X)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 3132623 167.4    5709258 305.0  5709258 305.0
Vcells 5863450  44.8   22267496 169.9 56666022 432.4
> rowStats <- microbenchmark(rowMads = rowMads(X, na.rm = FALSE), rowMads2 = rowMads2(X, na.rm = FALSE), 
+     `apply+mad` = apply(X, MARGIN = 1L, FUN = mad, na.rm = FALSE), unit = "ms")

Table: Benchmarking of colMads(), colMads2() and apply+mad() 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 colMads 3.301402 3.336935 3.441625 3.369768 3.445602 4.353206
2 colMads2 3.404427 3.462488 3.839495 3.513167 3.626749 15.573139
3 apply+mad 9.550131 9.723353 10.202366 9.851900 10.085211 19.316719
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 colMads2 1.031206 1.037625 1.115605 1.042554 1.052573 3.577395
3 apply+mad 2.892750 2.913858 2.964403 2.923614 2.926981 4.437355

Table: Benchmarking of rowMads(), rowMads2() and apply+mad() 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 rowMads 3.330468 3.348110 3.425502 3.385632 3.442791 4.230056
2 rowMads2 3.452394 3.482464 3.672875 3.536372 3.600909 10.834396
3 apply+mad 9.656374 9.694603 10.244143 9.776920 9.922432 19.092269
expr min lq mean median uq max
1 rowMads 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
2 rowMads2 1.036609 1.040129 1.072215 1.044523 1.045927 2.561289
3 apply+mad 2.899405 2.895545 2.990553 2.887769 2.882090 4.513479

Figure: Benchmarking of colMads(), colMads2() and apply+mad() on double+1000x100 data as well as rowMads(), rowMads2() and apply+mad() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colMads() and rowMads() 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 colMads 3.301402 3.336935 3.441625 3.369768 3.445602 4.353206
2 rowMads 3.330468 3.348110 3.425502 3.385632 3.442791 4.230056
expr min lq mean median uq max
1 colMads 1.000000 1.000000 1.0000000 1.000000 1.000000 1.0000000
2 rowMads 1.008804 1.003349 0.9953151 1.004708 0.999184 0.9717105

Figure: Benchmarking of colMads() and rowMads() on double+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

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

Matrix products: default
BLAS:   /home/hb/software/R-devel/R-3-6-branch/lib/R/lib/libRblas.so
LAPACK: /home/hb/software/R-devel/R-3-6-branch/lib/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] microbenchmark_1.4-6    matrixStats_0.55.0-9000 ggplot2_3.2.1          
[4] knitr_1.24              R.devices_2.16.0        R.utils_2.9.0          
[7] R.oo_1.22.0             R.methodsS3_1.7.1       history_0.0.0-9002     

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

Total processing time was 1.32 mins.

Reproducibility

To reproduce this report, do:

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

Copyright Henrik Bengtsson. Last updated on 2019-09-10 20:43:11 (-0700 UTC). Powered by RSP.

<script> var link = document.createElement('link'); link.rel = 'icon'; link.href = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAAA21BMVEUAAAAAAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8AAP8BAf4CAv0DA/wdHeIeHuEfH+AgIN8hId4lJdomJtknJ9g+PsE/P8BAQL9yco10dIt1dYp3d4h4eIeVlWqWlmmXl2iYmGeZmWabm2Tn5xjo6Bfp6Rb39wj4+Af//wA2M9hbAAAASXRSTlMAAQIJCgsMJSYnKD4/QGRlZmhpamtsbautrrCxuru8y8zN5ebn6Pn6+///////////////////////////////////////////LsUNcQAAAS9JREFUOI29k21XgkAQhVcFytdSMqMETU26UVqGmpaiFbL//xc1cAhhwVNf6n5i5z67M2dmYOyfJZUqlVLhkKucG7cgmUZTybDz6g0iDeq51PUr37Ds2cy2/C9NeES5puDjxuUk1xnToZsg8pfA3avHQ3lLIi7iWRrkv/OYtkScxBIMgDee0ALoyxHQBJ68JLCjOtQIMIANF7QG9G9fNnHvisCHBVMKgSJgiz7nE+AoBKrAPA3MgepvgR9TSCasrCKH0eB1wBGBFdCO+nAGjMVGPcQb5bd6mQRegN6+1axOs9nGfYcCtfi4NQosdtH7dB+txFIpXQqN1p9B/asRHToyS0jRgpV7nk4nwcq1BJ+x3Gl/v7S9Wmpp/aGquum7w3ZDyrADFYrl8vHBH+ev9AUASW1dmU4h4wAAAABJRU5ErkJggg==" document.getElementsByTagName('head')[0].appendChild(link); </script>
Add a custom footer
⚠️ **GitHub.com Fallback** ⚠️