Custom Functions and Nested Variables

Question

I'm still relatively new to the tidyverse and PURRR in particular. I'm working with a large dataset that combines nested and non-nested variables, and I want to use a mix of both to run a custom function. I've created dummy functions to show you what I mean, and attached the dput of an example dataset below (my apologies for the length - the nested data makes it quite long).

I have created a dummy function as well, which I attach here.

dummy.fn <- function(data, Temperature, P) {
  library(tidyverse)
  output <- sum(data$Mean_Flux*(Temperature*P))*1e-20
  return(output)
}

For each line, this function should take as input: the nested variable (here called Flux.data), and the corresponding temperature and pressure columns of the dummy dataset, and then perform this simple algebra. My function call appears as follows:

test <- mutate(dummy.df, example = dummy.fn(data = data,
                                            Temperature = Temperature,
                                            P = P))

For reasons I'm not clear on, this operation fails, as the new variable example is 0 in every line. In contrast, if I loop through the data and enter the variables piecewise, I get

realfake = numeric(length = 6)
for(ii in 1:6) {
  realfake[ii] = dummy.fn(data = dummy.df$data[[ii]], 
                          Temperature = dummy.df$Temperature[ii], 
                          P = dummy.df$P[ii])
}

> print(realfake)
[1] 7.822469 7.623985 3.539980 5.221917 3.287594 3.233342

I assume that I should be using something like map here instead of my simple mutate, but I'm not sure how to deal with the combination of nested and non-nested values.

Data

structure(list(season = c(3, 3, 1, 3, 3, 1), Altitude = c("MT",  "MT", "UT", "MT", "LT", "UT"), UTC_Time = structure(c(1470534960,  1470535080, 1486178400, 1470535200, 1470535320, 1486178520), class = c("POSIXct",  "POSIXt"), tzone = ""), P = c(589.97, 694.47, 376.68,
793.04, 
865.04, 352.25), Temperature = c(275.813611111111, 283.139444444444, 
257.765, 289.519444444444, 292.139722222222, 253.445555555556 ), data = list(structure(list(Wavelength = c(297, 298, 299, 300,  301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313,  314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326,  327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(346786701963.935, 
626061924535.969, 1126633726569.86, 1918338536593.65, 3218638104811.68, 
5449524002604.49, 9484898064254.88, 13820461435289.1, 18789920323010.3, 
23457046439017.9, 31321356743728.2, 39269949244536.7, 43130338001947.1,  52184025292495, 70006345247561.1, 80548638622151.5, 90086921275887.9, 
98153308839011.1, 103401175230574, 107092723165457, 126891646052471,  133856552389051, 141642783823426, 156803183198368, 163612559652001,  159862508775884, 157167983622353, 176116682634322, 199921821427841,  235385911421404, 249676715927087, 244363646529101, 263234566112574,  281333010457433, 265537025544415, 264427548933606, 261716337991127,  261346751545484, 270936371232163)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")), structure(list(Wavelength = c(297, 298, 299,  300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,  313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325,  326, 327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(272023148470.228, 
518233644515.949, 916010533767.96, 1565942149268.24, 2636296423894.42, 
4458908538328.54, 7732598175165.71, 11246928069805.3, 15264354612772.2, 
19019479942303.6, 25375401979585.9, 31737542925788.3, 34887526297163.1, 
42129119893975.9, 56514483527965.4, 64937287417956.8, 72609089666033.8, 
79076572477421.1, 83265097132731.9, 86245421746183.5, 102168967093555,  107683345120850, 114048659026611, 126228311354447, 131687088369977,  128667372125788, 126492872763247, 141800182905113, 161124838735026,  189719525471917, 201314850025731, 197022084150577, 212449864588280,  227084864091805, 214363865211836, 213506487998880, 211375083902905,  211159274561343, 218982868415835)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")), structure(list(Wavelength = c(297, 298, 299,  300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,  313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325,  326, 327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(410394304909.122, 
684669539810.892, 1223266694384.72, 1963557186902.24, 3173415679630.32, 
5225375319418.26, 8770197550604.42, 12564560749371.3, 16746223663975.8, 
20302741354880.3, 27273955562084.5, 32927242244373.8, 35498107396829.6, 
42892000512231.7, 56905187544496.6, 64359473555914.1, 71007069308199.5, 
76999513227547.8, 80793102650481, 83594711964789.9, 98448759760262.7,  102672040789990, 108260963181122, 120081240175307, 124208435953767,  120527913821372, 118657919972015, 132436584508003, 150262974200508,  176158060683751, 186058420013994, 181901206957521, 195971524806382,  208355140062145, 197630550381303, 196148678371193, 193073080285136,  192828255132408, 198899715679545)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")), structure(list(Wavelength = c(297, 298, 299,  300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,  313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325,  326, 327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(163980796868.381, 
319993777135.726, 565806448670.383, 952234326748.916, 1587959819258.22, 
2695812851544.62, 4651974671696.67, 6732838672280.01, 9108259191538.7, 
11334161396213.6, 15048347042419.1, 18807745701333.9, 20611119116971.8, 
24871101998997.2, 33301542059082.8, 38225550132546.5, 42687331288244.4, 
46437737119994.5, 48840611130603.3, 50597339661460.7, 59887648834929.7, 
63080591122608.5, 66751509164580.2, 73899774592187.1, 77054311628771.3, 
75283217576979.8, 74028011494075.6, 82984607314213.3, 94322313089456.9,  111107417274732, 117870238469697, 115397830776186, 124535437696499,  133132448331574, 125690170404882, 125251095649920, 124004362932209,  123933924746482, 128592585854440)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")), structure(list(Wavelength = c(297, 298, 299,  300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,  313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(100712940929.025, 
187783570779.657, 332125026989.09, 554468090233.077, 932816905860.644, 
1575341294540.1, 2706869628518.39, 3908049344198.36, 5283370151094.56, 
6563640428537.23, 8678472367958.32, 10859984329782.8, 11898351315247.2, 
14337012448330.9, 19158990854357.7, 21962913406403.8, 24514874805940.1, 
26658483263231.1, 28016129393510.1, 29012144950518.3, 34305093958882.3, 
36130169496521.6, 38221623366245, 42277066410473.2, 44074263806170.4, 
43065614437130.6, 42344345338240, 47429742792220.1, 53889760428024.4, 
63448640636975.8, 67287380731121.4, 65912251585506.6, 71093334728612.9, 
75986858424911.5, 71770877409023.9, 71485099232950.9, 70767556160329.4, 
70786293432652.9, 73403698511355.1)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")), structure(list(Wavelength = c(297, 298, 299,  300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,  313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325,  326, 327, 328, 329, 330, 331, 332, 333, 334, 335), Mean_Flux = c(415506831223.724, 
714394289371.61, 1265107560445.88, 2034229344334.13, 3270851868692.03, 
5361224636143.47, 8958831478561.96, 12807617463343.5, 17006209162760.4, 
20575733539097.5, 27539627927144.3, 33190657526915.2, 35712780965107.8, 
43093802584092.1, 57086484715867.1, 64512835082552.1, 71078004042059.9, 
76988088838629.3, 80651447645443, 83387384453277.3, 98085877600768.4,  102231449647244, 107777014668132, 119426399476690, 123467890246235,  119788043869974, 117832629287271, 131435264029990, 149046662645431,  174675091816222, 184386770145488, 180228275791829, 194042071016215,  206195905093624, 195517624876122, 194014778014755, 190866480268480,  190535424697236, 196524281939270)), .Names = c("Wavelength",  "Mean_Flux"), row.names = c(NA, -39L), class = c("tbl_df", "tbl",  "data.frame")))), row.names = c(NA, -6L), class = c("grouped_df",  "tbl_df", "tbl", "data.frame"), vars = c("season", "Altitude" ), drop
= TRUE, indices = list(c(2L, 5L), 4L, c(0L, 1L, 3L)), group_sizes = c(2L,  1L, 3L), biggest_group_size = 3L, labels = structure(list(season = c(1,  3, 3), Altitude = c("UT", "LT", "MT")), row.names = c(NA, -3L ), class = "data.frame", vars = c("season", "Altitude"), drop = TRUE, .Names = c("season",  "Altitude")), .Names = c("season", "Altitude", "UTC_Time", "P",  "Temperature", "data"))

www · Accepted Answer · 2019-02-19 02:07:10Z

1

First, no need to library(tidyverse)every time when you called your dummy.fn, it is a waste of computing resources, especially the function you wrote does not use any functions from the tidyverse. The following will work.

dummy.fn <- function(data, Temperature, P){
  output <- sum(data$Mean_Flux*(Temperature*P))*1e-20
  return(output)
}

Here is the code using the pmap_dbl from the purrr package to mimic your for loop. Notice that in your data element, there are only four data frames. To match the length of data, I subset the other two elements to only have length of four.

library(tidyverse)

pmap_dbl(list(data = dummy.df$data, Temperature = dummy.df$Temperature[1:4], P = dummy.df$P[1:4]),
         dummy.fn)

# [1] 7.8224688 7.6239845 3.5399800 0.7506653

edited Feb 19, 2019 at 2:07

answered Feb 19, 2019 at 1:56

www

39.3k12 gold badges52 silver badges93 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Custom Functions and Nested Variables

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related