ockovani.r.hacks.R

library(plyr)
data <- read.csv("~/Downloads/ockovaci-mista.csv")
data$datum <- as.Date(data$datum)
data$week <- strftime(data$datum, format = "%Y-%V")
prvni.davka <- data[ data$poradi_davky == 1,]
by.week <- ddply(prvni.davka, .(week), .fun = function(part) {
  data.frame(cnt = nrow(part))
})
add <- by.week[by.week$week == "2021-53", "cnt"]
by.week[by.week$week == "2020-53", "cnt"] = add + by.week[by.week$week == "2020-53", "cnt"] 
by.week <- by.week[by.week$week != "2021-53", ]
by.week$ord <- c()
for (i in 1:nrow(by.week)) {
  by.week$ord[i] <- i
}
linearMod <- lm(cnt ~ ord, by.week)
weeks <- data.frame(
  ord = 16:53
)
preds <- data.frame(
  ord = weeks$ord,
  cnt = predict(linearMod, weeks)
)
plot(
  by.week$ord, by.week$cnt, 
  xlim=c(min(by.week$ord), max(preds$ord)),
  ylim=c(min(by.week$cnt), max(preds$cnt)),
)
points(preds$ord, preds$cnt, col="red")
done <- sum(by.week$cnt)
for (i in 1:nrow(preds)) {
  pred <- sum(preds[1:i, "cnt"])
  cat(paste(preds[i, "ord"], "week->", format(done + pred, big.mark=" "), "\n"))
}

output.txt

16 week-> 1 406 279 
17 week-> 1 590 883 
18 week-> 1 786 864 
19 week-> 1 994 224 
20 week-> 2 212 961 
21 week-> 2 443 076 
22 week-> 2 684 569 
23 week-> 2 937 440 
24 week-> 3 201 688 
25 week-> 3 477 314 
26 week-> 3 764 318 
27 week-> 4 062 700 
28 week-> 4 372 460 
29 week-> 4 693 597 
30 week-> 5 026 113 
31 week-> 5 370 006 
32 week-> 5 725 277 
33 week-> 6 091 925 
34 week-> 6 469 952 
35 week-> 6 859 356 
36 week-> 7 260 138 
37 week-> 7 672 298 
38 week-> 8 095 836 
39 week-> 8 530 752 
40 week-> 8 977 045 
41 week-> 9 434 716 
42 week-> 9 903 765 
43 week-> 10 384 192 
44 week-> 10 875 996 
45 week-> 11 379 179 
46 week-> 11 893 739 
47 week-> 12 419 677 
48 week-> 12 956 993 
49 week-> 13 505 686 
50 week-> 14 065 758 
51 week-> 14 637 207 
52 week-> 15 220 034 
53 week-> 15 814 239