R 微基准警告 - 无法测量 x 评估的正执行时间
R microbenchmark warning - Could not measure a positive execution time for x evaluations
我正在开发一个绘制动物园对象的函数，因此我正在使用 microbenchmark.
对各种选项的执行速度进行基准测试
但是，每次我 运行 microbenchmark 我都会收到以下错误：
Warning message:
In microbenchmark::microbenchmark(overlapping_base, overlapping_lattice, :
Could not measure a positive execution time for 2524 evaluations.
每次我 运行 代码时，评估值都在变化。
我的猜测是基准测试无法完成。这意味着每次我 运行 和 microbenchmark 我都会得到非常不同的响应（这极大地改变了执行速度）。因此，我无法确定哪种绘图方法最快。
我使用的代码：
library(lattice)
library(ggplot2)
library(zoo)
z <- structure(c(-0.02838132, -0.02642141, -0.02511601, -0.02443906,
-0.02414024, -0.02417388, -0.02451562, -0.02393946, -0.02242496,
-0.02104852, 0.01739089, 0.01981665, 0.02210309, 0.02376982,
0.02496479, 0.02538433, 0.02558674, 0.02642987, 0.02848319, 0.0305888,
-0.02174604, -0.01987625, -0.01885374, -0.01837707, -0.01820114,
-0.01826306, -0.01862707, -0.01826593, -0.01710169, -0.01580846,
-0.00849795, -0.00590564, -0.00353068, -0.00165086, -0.00038422,
0.00032302, 0.00035668, 0.00161519, 0.00415983, 0.00657286, 0.00965842,
0.01212343, 0.01442211, 0.01579477, 0.01667243, 0.01666898, 0.01611297,
0.01692785, 0.01928281, 0.02143322, -0.0194464, -0.01757103,
-0.0158406, -0.01457289, -0.01380771, -0.0135324, -0.0140401,
-0.01340732, -0.01169943, -0.00988943, -0.02916019, -0.02673978,
-0.02466732, -0.0237339, -0.02308799, -0.02317864, -0.02376145,
-0.0227629, -0.02004088, -0.01753194, -0.05127982, -0.04929435,
-0.04880177, -0.04996632, -0.05004232, -0.0493009, -0.05101994,
-0.05113115, -0.04892411, -0.0475149, -0.07164312, -0.0686314,
-0.0660294, -0.06348602, -0.06356825, -0.06302109, -0.06276327,
-0.06209531, -0.05954085, -0.05830172, 0.01311737, 0.0148998,
0.01550045, 0.01542315, 0.01480778, 0.01336875, 0.01106039, 0.01161289,
0.01314323, 0.0142203, -0.07079136, -0.06946734, -0.06746012,
-0.06623026, -0.06813571, -0.06920641, -0.07168502, -0.0710586,
-0.06919208, -0.06819284, -0.04583223, -0.04687478, -0.04720935,
-0.04925846, -0.04846798, -0.05010275, -0.04891061, -0.04988978,
-0.04763314, -0.04486417, 0.01508285, 0.01771387, 0.01775548,
0.0186012, 0.01964036, 0.0206486, 0.02302578, 0.02525439, 0.02973214,
0.0317942, 0.00488448, 0.00831908, 0.01058987, 0.01139279, 0.01237221,
0.01294144, 0.01263533, 0.01327397, 0.01563062, 0.01794102, -0.00091997,
-0.00010884, 0.00179444, 0.00415676, 0.00597374, 0.00498312,
0.00294308, 0.00433847, 0.00660325, 0.00808107, 0.01277616, 0.0147175,
0.01651502, 0.01767393, 0.0183272, 0.01811757, 0.01746553, 0.01833143,
0.02001129, 0.02195872, 0.03273455, 0.03516022, 0.03719699, 0.03895484,
0.04029506, 0.04113609, 0.04134692, 0.04248674, 0.0445687, 0.04700408,
-0.03783289, -0.03577733, -0.03476676, -0.03349859, -0.03263628,
-0.03324435, -0.0336928, -0.03278793, -0.03165878, -0.03065102,
0.02888535, 0.03179749, 0.03449393, 0.03605427, 0.037507, 0.03839276,
0.03837656, 0.03956103, 0.04147407, 0.0435806, -0.0104104, -0.00884235,
-0.00788977, -0.007072, -0.00677401, -0.00643356, -0.0062355,
-0.00593316, -0.00524193, -0.0045362, -0.07363815, -0.0713515,
-0.06969982, -0.06867965, -0.0681607, -0.06815293, -0.06900493,
-0.068134, -0.06649919, -0.06501484, 0.02291233, 0.02638201,
0.02907269, 0.03113551, 0.03215457, 0.03223916, 0.03193051, 0.03326608,
0.03605787, 0.03868543, -0.14471737, -0.14382875, -0.1425845,
-0.14208004, -0.14131485, -0.14088634, -0.14169012, -0.1417378,
-0.14083509, -0.13993506, -0.04463136, -0.04119767, -0.03801833,
-0.03594387, -0.03533695, -0.03557779, -0.03625033, -0.03449043,
-0.03091173, -0.0278551), .Dim = c(10L, 24L), .Dimnames = list(
NULL, c("CH1", "CH2", "CH3", "CH4", "CH5", "CH6", "CH7",
"CH8", "CH9", "CH10", "CH11", "CH12", "CH13", "CH14", "CH15",
"CH16", "CH17", "CH18", "CH19", "CH20", "CH21", "CH22", "CH23",
"CH24")), index = 1:10, class = "zoo")
# benchmark plotting operations
overlapping_base <- plot(z, screens = 1, col = 1:ncol(z))
overlapping_lattice <- xyplot(z, screens = 1, col = 1:ncol(z))
overlapping_ggplot2 <- autoplot(z, facets = NULL) + aes(linetype = NULL)
microbenchmark::microbenchmark(overlapping_base, overlapping_lattice, overlapping_ggplot2, times = 1000)
为什么我会收到该警告或如何修复它有什么想法吗？
如果您以极快的执行速度对某项进行基准测试，则此警告很典型。如果您查看基准测试，您会看到 0 ns 的时序。请注意，在基准测试运行期间没有实际绘图发生。
执行 overlapping_base 以查看您实际进行的基准测试。请注意，如果网格图未包装在 print.
中，则不会在 non-interactive 中绘制
正在修复您的代码：
# benchmark plotting operations
overlapping_base <- function() plot(z, screens = 1, col = 1:ncol(z))
overlapping_lattice <- function() print(xyplot(z, screens = 1, col = 1:ncol(z)))
overlapping_ggplot2 <- function() print(autoplot(z, facets = NULL) + aes(linetype = NULL))
microbenchmark::microbenchmark(overlapping_base(),
overlapping_lattice(),
overlapping_ggplot2(), times = 10)
#Unit: milliseconds
# expr min lq mean median uq max neval cld
# overlapping_base() 13.63389 19.36380 100.2684 152.1086 154.7808 160.4726 10 a
# overlapping_lattice() 49.70754 51.79813 123.7162 125.9431 192.6951 204.0190 10 a
# overlapping_ggplot2() 238.17771 249.62242 342.9059 388.5780 411.9674 420.9687 10 b
我正在开发一个绘制动物园对象的函数，因此我正在使用 microbenchmark.
对各种选项的执行速度进行基准测试但是，每次我 运行 microbenchmark 我都会收到以下错误：
Warning message:
In microbenchmark::microbenchmark(overlapping_base, overlapping_lattice, : Could not measure a positive execution time for 2524 evaluations.
每次我 运行 代码时，评估值都在变化。
我的猜测是基准测试无法完成。这意味着每次我 运行 和 microbenchmark 我都会得到非常不同的响应（这极大地改变了执行速度）。因此，我无法确定哪种绘图方法最快。
我使用的代码：
library(lattice)
library(ggplot2)
library(zoo)
z <- structure(c(-0.02838132, -0.02642141, -0.02511601, -0.02443906,
-0.02414024, -0.02417388, -0.02451562, -0.02393946, -0.02242496,
-0.02104852, 0.01739089, 0.01981665, 0.02210309, 0.02376982,
0.02496479, 0.02538433, 0.02558674, 0.02642987, 0.02848319, 0.0305888,
-0.02174604, -0.01987625, -0.01885374, -0.01837707, -0.01820114,
-0.01826306, -0.01862707, -0.01826593, -0.01710169, -0.01580846,
-0.00849795, -0.00590564, -0.00353068, -0.00165086, -0.00038422,
0.00032302, 0.00035668, 0.00161519, 0.00415983, 0.00657286, 0.00965842,
0.01212343, 0.01442211, 0.01579477, 0.01667243, 0.01666898, 0.01611297,
0.01692785, 0.01928281, 0.02143322, -0.0194464, -0.01757103,
-0.0158406, -0.01457289, -0.01380771, -0.0135324, -0.0140401,
-0.01340732, -0.01169943, -0.00988943, -0.02916019, -0.02673978,
-0.02466732, -0.0237339, -0.02308799, -0.02317864, -0.02376145,
-0.0227629, -0.02004088, -0.01753194, -0.05127982, -0.04929435,
-0.04880177, -0.04996632, -0.05004232, -0.0493009, -0.05101994,
-0.05113115, -0.04892411, -0.0475149, -0.07164312, -0.0686314,
-0.0660294, -0.06348602, -0.06356825, -0.06302109, -0.06276327,
-0.06209531, -0.05954085, -0.05830172, 0.01311737, 0.0148998,
0.01550045, 0.01542315, 0.01480778, 0.01336875, 0.01106039, 0.01161289,
0.01314323, 0.0142203, -0.07079136, -0.06946734, -0.06746012,
-0.06623026, -0.06813571, -0.06920641, -0.07168502, -0.0710586,
-0.06919208, -0.06819284, -0.04583223, -0.04687478, -0.04720935,
-0.04925846, -0.04846798, -0.05010275, -0.04891061, -0.04988978,
-0.04763314, -0.04486417, 0.01508285, 0.01771387, 0.01775548,
0.0186012, 0.01964036, 0.0206486, 0.02302578, 0.02525439, 0.02973214,
0.0317942, 0.00488448, 0.00831908, 0.01058987, 0.01139279, 0.01237221,
0.01294144, 0.01263533, 0.01327397, 0.01563062, 0.01794102, -0.00091997,
-0.00010884, 0.00179444, 0.00415676, 0.00597374, 0.00498312,
0.00294308, 0.00433847, 0.00660325, 0.00808107, 0.01277616, 0.0147175,
0.01651502, 0.01767393, 0.0183272, 0.01811757, 0.01746553, 0.01833143,
0.02001129, 0.02195872, 0.03273455, 0.03516022, 0.03719699, 0.03895484,
0.04029506, 0.04113609, 0.04134692, 0.04248674, 0.0445687, 0.04700408,
-0.03783289, -0.03577733, -0.03476676, -0.03349859, -0.03263628,
-0.03324435, -0.0336928, -0.03278793, -0.03165878, -0.03065102,
0.02888535, 0.03179749, 0.03449393, 0.03605427, 0.037507, 0.03839276,
0.03837656, 0.03956103, 0.04147407, 0.0435806, -0.0104104, -0.00884235,
-0.00788977, -0.007072, -0.00677401, -0.00643356, -0.0062355,
-0.00593316, -0.00524193, -0.0045362, -0.07363815, -0.0713515,
-0.06969982, -0.06867965, -0.0681607, -0.06815293, -0.06900493,
-0.068134, -0.06649919, -0.06501484, 0.02291233, 0.02638201,
0.02907269, 0.03113551, 0.03215457, 0.03223916, 0.03193051, 0.03326608,
0.03605787, 0.03868543, -0.14471737, -0.14382875, -0.1425845,
-0.14208004, -0.14131485, -0.14088634, -0.14169012, -0.1417378,
-0.14083509, -0.13993506, -0.04463136, -0.04119767, -0.03801833,
-0.03594387, -0.03533695, -0.03557779, -0.03625033, -0.03449043,
-0.03091173, -0.0278551), .Dim = c(10L, 24L), .Dimnames = list(
NULL, c("CH1", "CH2", "CH3", "CH4", "CH5", "CH6", "CH7",
"CH8", "CH9", "CH10", "CH11", "CH12", "CH13", "CH14", "CH15",
"CH16", "CH17", "CH18", "CH19", "CH20", "CH21", "CH22", "CH23",
"CH24")), index = 1:10, class = "zoo")
# benchmark plotting operations
overlapping_base <- plot(z, screens = 1, col = 1:ncol(z))
overlapping_lattice <- xyplot(z, screens = 1, col = 1:ncol(z))
overlapping_ggplot2 <- autoplot(z, facets = NULL) + aes(linetype = NULL)
microbenchmark::microbenchmark(overlapping_base, overlapping_lattice, overlapping_ggplot2, times = 1000)
为什么我会收到该警告或如何修复它有什么想法吗？
如果您以极快的执行速度对某项进行基准测试，则此警告很典型。如果您查看基准测试，您会看到 0 ns 的时序。请注意，在基准测试运行期间没有实际绘图发生。
执行 overlapping_base 以查看您实际进行的基准测试。请注意，如果网格图未包装在 print.
正在修复您的代码：
# benchmark plotting operations
overlapping_base <- function() plot(z, screens = 1, col = 1:ncol(z))
overlapping_lattice <- function() print(xyplot(z, screens = 1, col = 1:ncol(z)))
overlapping_ggplot2 <- function() print(autoplot(z, facets = NULL) + aes(linetype = NULL))
microbenchmark::microbenchmark(overlapping_base(),
overlapping_lattice(),
overlapping_ggplot2(), times = 10)
#Unit: milliseconds
# expr min lq mean median uq max neval cld
# overlapping_base() 13.63389 19.36380 100.2684 152.1086 154.7808 160.4726 10 a
# overlapping_lattice() 49.70754 51.79813 123.7162 125.9431 192.6951 204.0190 10 a
# overlapping_ggplot2() 238.17771 249.62242 342.9059 388.5780 411.9674 420.9687 10 b