Error in UseMethod("predict") :
  no applicable method for 'predict' applied to an object of class "character"
In addition: Warning messages:
1: In `[<-.factor`(`*tmp*`, n, value = "A1") :
  invalid factor level, NA generated
2: In `[<-.factor`(`*tmp*`, n, value = "A2") :
  invalid factor level, NA generated
3: In `[<-.factor`(`*tmp*`, n, value = "A3") :
  invalid factor level, NA generated
4: In min(data_n[data_n > 0]) :
  no non-missing arguments to min; returning Inf
5: In `[<-.factor`(`*tmp*`, n, value = "A4") :
  invalid factor level, NA generated

library(growthcurver)
d <- growthdata
file<-"C:/Users/bedro/Downloads/growth curve.csv"
d <- read.table(file, sep=",", header=TRUE, stringsAsFactors=FALSE)
#d$time <- d$time / 60
num_analyses <- length(names(d)) - 1
d_gc <- data.frame(sample = character(num_analyses),k = numeric(num_analyses),n0  = numeric(num_analyses),r = numeric(num_analyses),t_mid = numeric(num_analyses),t_gen = numeric(num_analyses),auc_l = numeric(num_analyses),auc_e = numeric(num_analyses), sigma = numeric(num_analyses),tringsAsFactors = FALSE)
par(mfcol = c(8,12))
par(mar = c(0.25,0.25,0.25,0.25))
trim_at_time <- 10
y_lim_max <- max(d[,setdiff(names(d), "time")]) - min(d[,setdiff(names(d), "time")])
n <- 1
pdf("C:/Desktop/Users/bedro/Desktop/growthcurverplot2.pdf", height = 8.5, width = 11)
par(mfcol = c(8,12))
par(mar = c(0.25,0.25,0.25,0.25))
y_lim_max <- max(d[,setdiff(names(d), "time")]) - min(d[,setdiff(names(d), "time")])
n <- 1    # keeps track of the current row in the output data frame
for (col_name in names(d)) {
   # Don't process the column called "time".
   # It contains time and not absorbance data.
   if (col_name != "time") {

     # Create a temporary data frame that contains just the time and current col
     d_loop <- d[, c("time", col_name)]

     # Do the background correction.
     # Background correction option 1: subtract the minimum value in a column
     #                                 from all measurements in that column
         min_value <- min(d_loop[, col_name])
     d_loop[, col_name] <- d_loop[, col_name] - min_value
     # Background correction option 2: subtract the mean value of blank wells
     #                                 over the course the experiment
     #                                 (Replace B2, D8, G11 with the column
     #                                  names of your media-only wells)
     #d$blank <- apply(d[, c("B2", "D8", "G11")], 1, mean)
     #d$A1 <- d$A1 - d$blank

     # Now, call Growthcurver to calculate the metrics using SummarizeGrowth
     gc_fit <- SummarizeGrowth(data_t = d_loop[, "time"],
                               data_n = d_loop[, col_name],
                               t_trim = trim_at_time,
                               bg_correct = "none")

     # Now, add the metrics from this column to the next row (n) in the
     # output data frame, and increment the row counter (n)
     d_gc$sample[n] <- col_name
     d_gc[n, 2:9] <- c(gc_fit$vals$k,
                       gc_fit$vals$n0,
                       gc_fit$vals$r,
                       gc_fit$vals$t_mid,
                       gc_fit$vals$t_gen,
                       gc_fit$vals$auc_l,
                       gc_fit$vals$auc_e,
                       gc_fit$vals$sigma)
     n <- n + 1

     # Finally, plot the raw data and the fitted curve
     # Here, I'll just print some of the data points to keep the file size smaller
     n_obs <- length(gc_fit$data$t)
     idx_to_plot <- 1:20 / 20 * n_obs
     plot(gc_fit$data$t[idx_to_plot], gc_fit$data$N[idx_to_plot],
          pch = 20,
          xlim = c(0, trim_at_time),
          ylim = c(0, y_lim_max),
          cex = 0.6, xaxt = "n", yaxt = "n")
      text(x = trim_at_time / 4, y = y_lim_max, labels = col_name, pos = 1)

      lines(gc_fit$data$t, predict(gc_fit$model), col = "red")
   }
 }