Best practices

I have a dataframe called Profile_1_2 of data that is formatted like this:

Profile_number chl_z Cphyto_z
1 0.2 10
1 0.6 100
2 0.1 20
2 0.5 90

Profile_number denotes a group, each profile contains 200 rows. I implemented a model from a python script using reticulate

for (i in c("tidyverse","reticulate")) {
  if (!require(i, character.only = TRUE)) {
    install.packages(i, dependencies = TRUE)
    library(i, character.only = TRUE)}}

np <- import("numpy")

source_python('cbpm_argo.py')

cbmp_argo is function with input of 5 variables:

def cbpm_argo(chl_z,Cphyto_z,irr,dayOfYear,lat):

and output of 7:

return pp_z,mu_z,par_z,prcnt_z,nutTempFunc_z,IgFunc_z,mzeu

when I use the function in a profile-wise basis it outputs a list of 7 elements without naming them:

Profile_1 <- Profile_1_2 %>% 
  filter(Profile_number == 1)

output <- cbpm_argo(Profile_1$chl_z,Profile_1$Cphyto_z,30,113,30)

summary(output)

     Length Class  Mode   
[1,] 200    -none- numeric
[2,] 200    -none- numeric
[3,] 200    -none- numeric
[4,] 200    -none- numeric
[5,] 200    -none- numeric
[6,] 200    -none- numeric
[7,]   1    -none- numeric

I want the model output to be added to the initial dataframe as additional columns, which I have done simply using square brackets, resulting in a dataframe with 9 columns as intended:

Profile_1_2 <- Profile_1_2 %>%
  mutate(.by = Profile_number,
         pp_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[1]],
         mu_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[2]],
         par_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[3]],
         prcnt_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[4]],
         nutTempFunc_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[5]],
         IgFunc_z = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[6]],
         mzeu = cbpm_argo(chl_z,Cphyto_z,30,113,30)[[7]]
         )

However this seems incredibly long winded and (I think) requires repeated use of the function for a single profile. Is there a more concise way of extracting the model output into the dataframe.

for those interested I am using a slightly altered version of the scripts listed by Lionel Arteaga https://zenodo.org/records/6599224#23.YqJZFC1h3uM

Improve this question