matrixStats.benchmarks

colLogSumExps() and rowLogSumExps() benchmarks on subsetted computation

This report benchmark the performance of colLogSumExps() and rowLogSumExps() on subsetted computation.

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 = "double")

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  5238049 279.8    7916910 422.9  7916910 422.9
Vcells 10108692  77.2   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
           used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells  5225312 279.1    7916910 422.9  7916910 422.9
Vcells 10066514  76.9   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.007115	0.0073195	0.0101712	0.0074750	0.0076295	0.236078
2	colLogSumExps(X, rows, cols)	0.007420	0.0077625	0.0083343	0.0079385	0.0081360	0.021503
3	colLogSumExps(X[rows, cols])	0.008110	0.0087055	0.0090248	0.0088625	0.0090620	0.018043

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	colLogSumExps(X, rows, cols)	1.042867	1.060523	0.8194032	1.062007	1.066387	0.0910843
3	colLogSumExps(X[rows, cols])	1.139845	1.189357	0.8872928	1.185619	1.187758	0.0764281

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.005605	0.005921	0.0062526	0.0060105	0.0061855	0.019133
2	rowLogSumExps(X, cols, rows)	0.006106	0.006320	0.0084117	0.0064450	0.0066540	0.181226
3	rowLogSumExps(X[cols, rows])	0.006688	0.007062	0.0079129	0.0072195	0.0074580	0.020867

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowLogSumExps(X, cols, rows)	1.089384	1.067387	1.345314	1.072290	1.075742	9.471907
3	rowLogSumExps(X[cols, rows])	1.193220	1.192704	1.265544	1.201148	1.205723	1.090629

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 10x10 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowLogSumExps_X_S	5.605	5.9210	6.25260	6.0105	6.1855	19.133
1	colLogSumExps_X_S	7.115	7.3195	10.17122	7.4750	7.6295	236.078

	expr	min	lq	mean	median	uq	max
2	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
1	colLogSumExps_X_S	1.269402	1.236193	1.626718	1.243657	1.233449	12.33879

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 5224037 279.0    7916910 422.9  7916910 422.9
Vcells 9737400  74.3   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5224031 279.0    7916910 422.9  7916910 422.9
Vcells 9747483  74.4   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.092659	0.0982260	0.1071898	0.1045055	0.1167345	0.137236
2	colLogSumExps(X, rows, cols)	0.104962	0.1109100	0.1214687	0.1166815	0.1329395	0.164765
3	colLogSumExps(X[rows, cols])	0.105212	0.1115215	0.1222706	0.1174300	0.1334405	0.189864

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colLogSumExps(X, rows, cols)	1.132777	1.129131	1.133212	1.116511	1.138819	1.200596
3	colLogSumExps(X[rows, cols])	1.135475	1.135356	1.140693	1.123673	1.143111	1.383485

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.092408	0.097453	0.1067365	0.1025205	0.1155995	0.134108
2	rowLogSumExps(X, cols, rows)	0.100721	0.107351	0.1175832	0.1146055	0.1246125	0.178518
3	rowLogSumExps(X[cols, rows])	0.105402	0.110651	0.1206870	0.1169775	0.1279125	0.163867

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowLogSumExps(X, cols, rows)	1.089960	1.101567	1.101621	1.117879	1.077967	1.331151
3	rowLogSumExps(X[cols, rows])	1.140616	1.135429	1.130700	1.141016	1.106514	1.221903

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 100x100 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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	rowLogSumExps_X_S	92.408	97.453	106.7365	102.5205	115.5995	134.108
1	colLogSumExps_X_S	92.659	98.226	107.1898	104.5055	116.7345	137.236

	expr	min	lq	mean	median	uq	max
2	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colLogSumExps_X_S	1.002716	1.007932	1.004247	1.019362	1.009818	1.023324

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 5224778 279.1    7916910 422.9  7916910 422.9
Vcells 9741449  74.4   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5224772 279.1    7916910 422.9  7916910 422.9
Vcells 9751532  74.4   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.081337	0.0878975	0.0918830	0.0917245	0.0936210	0.134037
3	colLogSumExps(X[rows, cols])	0.090477	0.0974915	0.1047085	0.1023490	0.1074920	0.166081
2	colLogSumExps(X, rows, cols)	0.091705	0.1020535	0.1065506	0.1056700	0.1096165	0.132534

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	colLogSumExps(X[rows, cols])	1.112372	1.109150	1.139585	1.115831	1.148161	1.2390683
2	colLogSumExps(X, rows, cols)	1.127470	1.161051	1.159633	1.152037	1.170854	0.9887867

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.089144	0.0936680	0.1016918	0.0985455	0.1077120	0.134465
2	rowLogSumExps(X, cols, rows)	0.095979	0.1018065	0.1105511	0.1076820	0.1195335	0.140813
3	rowLogSumExps(X[cols, rows])	0.105246	0.1092500	0.1196410	0.1164450	0.1283870	0.210126

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowLogSumExps(X, cols, rows)	1.076674	1.086887	1.087119	1.092713	1.109751	1.047209
3	rowLogSumExps(X[cols, rows])	1.180629	1.166354	1.176506	1.181637	1.191947	1.562682

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 1000x10 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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	colLogSumExps_X_S	81.337	87.8975	91.88296	91.7245	93.621	134.037
2	rowLogSumExps_X_S	89.144	93.6680	101.69176	98.5455	107.712	134.465

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.00000	1.000000	1.000000	1.000000	1.000000
2	rowLogSumExps_X_S	1.095983	1.06565	1.106753	1.074364	1.150511	1.003193

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 5224982 279.1    7916910 422.9  7916910 422.9
Vcells 9742375  74.4   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5224976 279.1    7916910 422.9  7916910 422.9
Vcells 9752458  74.5   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.105838	0.1169940	0.1325966	0.1280575	0.1437870	0.199453
3	colLogSumExps(X[rows, cols])	0.121829	0.1324750	0.1430677	0.1380670	0.1509265	0.210831
2	colLogSumExps(X, rows, cols)	0.119458	0.1310025	0.1445732	0.1394405	0.1508250	0.214062

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colLogSumExps(X[rows, cols])	1.151089	1.132323	1.078969	1.078164	1.049653	1.057046
2	colLogSumExps(X, rows, cols)	1.128687	1.119737	1.090323	1.088890	1.048947	1.073245

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.107426	0.1172715	0.1286174	0.1229725	0.1376535	0.169853
3	rowLogSumExps(X[cols, rows])	0.116121	0.1300935	0.1426984	0.1354825	0.1564030	0.179422
2	rowLogSumExps(X, cols, rows)	0.119500	0.1299565	0.1421029	0.1384615	0.1486940	0.214896

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.00000	1.000000	1.000000	1.000000
3	rowLogSumExps(X[cols, rows])	1.080939	1.109336	1.10948	1.101730	1.136208	1.056337
2	rowLogSumExps(X, cols, rows)	1.112394	1.108168	1.10485	1.125955	1.080205	1.265188

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 10x1000 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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	rowLogSumExps_X_S	107.426	117.2715	128.6174	122.9725	137.6535	169.853
1	colLogSumExps_X_S	105.838	116.9940	132.5966	128.0575	143.7870	199.453

	expr	min	lq	mean	median	uq	max
2	rowLogSumExps_X_S	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
1	colLogSumExps_X_S	0.9852177	0.9976337	1.030938	1.041351	1.044558	1.174268

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 5225195 279.1    7916910 422.9  7916910 422.9
Vcells 9787108  74.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5225189 279.1    7916910 422.9  7916910 422.9
Vcells 9887191  75.5   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.732019	0.8122615	0.8815236	0.908237	0.931889	1.290778
3	colLogSumExps(X[rows, cols])	0.824849	0.9041215	1.1203326	1.052028	1.148853	9.666148
2	colLogSumExps(X, rows, cols)	0.823153	0.9286710	1.0548352	1.101485	1.165216	1.408639

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
3	colLogSumExps(X[rows, cols])	1.126814	1.113092	1.270905	1.158319	1.232821	7.488622
2	colLogSumExps(X, rows, cols)	1.124497	1.143315	1.196605	1.212773	1.250381	1.091310

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.763864	0.8275575	1.010433	0.9387495	0.9760135	9.037269
3	rowLogSumExps(X[cols, rows])	0.864047	0.9813495	1.105827	1.1447305	1.1737865	1.552095
2	rowLogSumExps(X, cols, rows)	0.886787	1.0083610	1.112867	1.1510915	1.1868720	1.548171

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	rowLogSumExps(X[cols, rows])	1.131153	1.185838	1.094408	1.219421	1.202633	0.1717438
2	rowLogSumExps(X, cols, rows)	1.160923	1.218479	1.101376	1.226197	1.216041	0.1713096

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 100x1000 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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	colLogSumExps_X_S	732.019	812.2615	881.5236	908.2370	931.8890	1290.778
2	rowLogSumExps_X_S	763.864	827.5575	1010.4334	938.7495	976.0135	9037.269

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowLogSumExps_X_S	1.043503	1.018831	1.146235	1.033595	1.047349	7.001412

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 5225405 279.1    7916910 422.9  7916910 422.9
Vcells 9787933  74.7   33191153 253.3 53339345 407.0
> colStats <- microbenchmark(colLogSumExps_X_S = colLogSumExps(X_S, na.rm = FALSE), `colLogSumExps(X, rows, cols)` = colLogSumExps(X, 
+     rows = rows, cols = cols, na.rm = FALSE), `colLogSumExps(X[rows, cols])` = colLogSumExps(X[rows, 
+     cols], na.rm = FALSE), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5225399 279.1    7916910 422.9  7916910 422.9
Vcells 9888016  75.5   33191153 253.3 53339345 407.0
> rowStats <- microbenchmark(rowLogSumExps_X_S = rowLogSumExps(X_S, na.rm = FALSE), `rowLogSumExps(X, cols, rows)` = rowLogSumExps(X, 
+     rows = cols, cols = rows, na.rm = FALSE), `rowLogSumExps(X[cols, rows])` = rowLogSumExps(X[cols, 
+     rows], na.rm = FALSE), unit = "ms")

Table: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	0.678350	0.8307220	0.9314281	0.855903	0.8693755	8.881709
3	colLogSumExps(X[rows, cols])	0.782717	0.8763665	0.9949185	1.027352	1.0821845	1.706902
2	colLogSumExps(X, rows, cols)	0.810869	0.9047975	1.0430200	1.100774	1.1264285	1.288961

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
3	colLogSumExps(X[rows, cols])	1.153854	1.054946	1.068165	1.200314	1.244784	0.1921817
2	colLogSumExps(X, rows, cols)	1.195355	1.089170	1.119807	1.286097	1.295675	0.1451253

Table: Benchmarking of rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on 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	rowLogSumExps_X_S	0.714063	0.8154905	0.9628592	0.903113	0.9221925	9.277656
2	rowLogSumExps(X, cols, rows)	0.824174	0.9135890	1.0372638	1.095503	1.1237030	1.350762
3	rowLogSumExps(X[cols, rows])	0.827792	0.9571870	1.0576103	1.106101	1.1349945	1.556035

	expr	min	lq	mean	median	uq	max
1	rowLogSumExps_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowLogSumExps(X, cols, rows)	1.154203	1.120294	1.077275	1.213029	1.218512	0.1455930
3	rowLogSumExps(X[cols, rows])	1.159270	1.173756	1.098406	1.224765	1.230757	0.1677185

Figure: Benchmarking of colLogSumExps_X_S(), colLogSumExps(X, rows, cols)() and colLogSumExps(X[rows, cols])() on 1000x100 data as well as rowLogSumExps_X_S(), rowLogSumExps(X, cols, rows)() and rowLogSumExps(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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	colLogSumExps_X_S	678.350	830.7220	931.4281	855.903	869.3755	8881.709
2	rowLogSumExps_X_S	714.063	815.4905	962.8592	903.113	922.1925	9277.656

	expr	min	lq	mean	median	uq	max
1	colLogSumExps_X_S	1.000000	1.0000000	1.000000	1.000000	1.000000	1.00000
2	rowLogSumExps_X_S	1.052647	0.9816647	1.033745	1.055158	1.060753	1.04458

Figure: Benchmarking of colLogSumExps_X_S() and rowLogSumExps_X_S() on 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 13.05 secs.

Reproducibility

To reproduce this report, do:

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