Unable to run a simple example

[Ravua1992]

Hi,

I am new to using nlmxir and I trying to fit a one compartment model to this data and have been unsuccessful so far.
Changing the initial estimates is not helping.
Thank you!

library(tidyr)
library(dplyr)
library(nlmixr2)
library(xpose)
library(xpose.nlmixr2)
library(ggplot2)
my_table <- data.frame(
  ID = 1,
  DV = c(NA, NA, 321000, 867000, 1070000, 1250000, 1130000, 883000, 1230000, 1420000, 2190000, 2650000, 2850000, 2180000, 2070000, 1790000),
  Day.Nominal = c(NA, 1, 1, 1, 1, 1, 1, 1, 15, 22, 22, 22, 22, 22, 22, 22),
  Hour.Nominal = c(NA, 0, 6, 24, 48, 72, 120, 168, 0, 0, 6, 24, 48, 72, 120, 168),
  Time = c(0, 24, 30, 48, 72, 96, 144, 192, 360, 528, 534, 552, 576, 600, 648, 696),
  Amt = c(125,rep(0, 15)),
  II = c(168,rep(0, 15)),
  addl = c(3,rep(0, 15)),
  EVID = c(101 , 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0),
  CMT = c(1 , 2, 2, 2,2,2,2,2,2,2,2,2,2,2,2,2),
  MDV = c(1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0)
)

my_table$ID <- as.integer(my_table$ID)

my_table$DV <- as.numeric(my_table$DV)

my_table<-my_table %>% arrange(ID,Time) 


one.compartment <- function() {
  ini({
    tka <- 0.5 # Log Ka
    tcl <- 0.003 # Log Cl
    tv <- 0.005   # Log V
    eta.ka ~ 0.6
    eta.cl ~ 0
    eta.v ~ 0
    add.sd <- 500
  })
  model({
    ka <- exp(tka + eta.ka)
    cl <- exp(tcl + eta.cl)
    v <- exp(tv + eta.v)
    d/dt(depot) = -ka * depot
    d/dt(center) = ka * depot - cl / v * center
    cp = center / v
    cp ~ add(add.sd)
  })
}

fit <- nlmixr(one.compartment, my_table, est="focei",
              control=list(print=0))

nhandled error message: EE:lsoda -- at t = 0 and step size _rxC(h) = 2.73124e-11, the
 corrector convergence failed repeatedly or
 with fabs(_rxC(h)) = hmin
 @(lsoda.c:889
Error in .foceiFitInternal(.ret) : 
  On initial gradient evaluation, one or more parameters have a zero gradient
Change model, try different initial estimates or use outerOpt="bobyqa")
Error : Could not fit data
  On initial gradient evaluation, one or more parameters have a zero gradient
Change model, try different initial estimates or use outerOpt="bobyqa")
Error: Could not fit data
  On initial gradient evaluation, one or more parameters have a zero gradient
Change model, try different initial estimates or use outerOpt="bobyqa")

[mattfidler]

This is a nlmixr2 question,

The disucssion board is

http://github.com/nlmixr2/nlmixr2/discussions