Anomalize error: no date or datetime column found

Question

I have some data:

   dput(y)
structure(list(date = structure(c(1508803200, 1515456000, 1506384000, 
1501027200, 1503964800, 1517356800, 1519776000, 1511740800, 1511827200, 
1498521600, 1509408000, 1522627200, 1493078400, 1502064000, 1504051200, 
1504569600, 1504742400, 1512432000, 1515628800, 1520467200, 1493164800, 
1496102400, 1496620800, 1501459200, 1506988800, 1508889600, 1512864000, 
1515542400, 1515974400, 1517443200, 1520553600, 1520812800, 1494806400, 
1495756800, 1496188800, 1496707200, 1498608000, 1499040000, 1499299200, 
1501113600, 1501200000, 1502150400, 1504656000, 1507075200, 1508976000, 
1509235200, 1509494400, 1515369600, 1516752000, 1516838400, 1517961600, 
1518393600, 1519689600, 1522368000, 1522540800), class = c("POSIXct", 
"POSIXt")), word = c("hr", "hr", "hr", "hr", "hr", "hr", "hr", 
"hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", 
"hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", 
"hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", 
"hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", "hr", 
"hr", "hr", "hr", "hr"), n = c(22L, 16L, 14L, 8L, 8L, 6L, 6L, 
5L, 5L, 4L, 4L, 4L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L
)), row.names = c(NA, -55L), vars = "date", drop = TRUE, class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), .Names = c("date", "word", "n"
), indices = list(12L, 20L, 32L, 33L, 21L, 34L, 22L, 35L, 9L, 
    36L, 37L, 38L, 3L, 39L, 40L, 23L, 13L, 41L, 4L, 14L, 15L, 
    42L, 16L, 2L, 24L, 43L, 0L, 25L, 44L, 45L, 10L, 46L, 7L, 
    8L, 17L, 26L, 47L, 1L, 27L, 18L, 28L, 48L, 49L, 5L, 29L, 
    50L, 51L, 52L, 6L, 19L, 30L, 31L, 53L, 54L, 11L), group_sizes = c(1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 1L, 1L, 1L), biggest_group_size = 1L, labels = structure(list(
    date = structure(c(1493078400, 1493164800, 1494806400, 1495756800, 
    1496102400, 1496188800, 1496620800, 1496707200, 1498521600, 
    1498608000, 1499040000, 1499299200, 1501027200, 1501113600, 
    1501200000, 1501459200, 1502064000, 1502150400, 1503964800, 
    1504051200, 1504569600, 1504656000, 1504742400, 1506384000, 
    1506988800, 1507075200, 1508803200, 1508889600, 1508976000, 
    1509235200, 1509408000, 1509494400, 1511740800, 1511827200, 
    1512432000, 1512864000, 1515369600, 1515456000, 1515542400, 
    1515628800, 1515974400, 1516752000, 1516838400, 1517356800, 
    1517443200, 1517961600, 1518393600, 1519689600, 1519776000, 
    1520467200, 1520553600, 1520812800, 1522368000, 1522540800, 
    1522627200), class = c("POSIXct", "POSIXt"))), row.names = c(NA, 
-55L), vars = "date", drop = TRUE, class = "data.frame", .Names = "date"))

and code:

y = y %>% 
  group_by(date) %>%
  unnest_tokens(word,comments) %>%
  anti_join(stop_words) %>%
  count(word,sort = T) 


  y %>% time_decompose(count,method='stl')%>%
    anomalize(remainder,method = 'iqr') %>%
    time_recompose()%>% plot_anomalies(time_recomposed = T,ncol=5,alpha_dots = 0.25)

except that I keep getting the following error:

Error in mutate_impl(.data, dots) : Evaluation error: Error in prep_tbl_time(): No date or datetime column found..

I followed the github tutorial for this and my columns and types match up to the example. But for some reason, it can't find the column date.

Which [line causes the error](http://idownvotedbecau.se/nodebugging/)? Can you `dput(head(y))` the beginning of your dataframe so we can [check where the error occurs for ourselves](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example)? Your syntax also suggests that the printed `y` is after some processing has been done, and is not the original dataframe from before the pipe starts (see the use of `rename` in line 1 which indicates that the original data has column `surveyDate` rather than `date`.) — Calum You, Apr 24 '18 at 18:28
What are you trying to accomplish with that first section? It's happening in there, but I can't pin it down. — Anonymous coward, Apr 24 '18 at 18:41
I had this issue recently, caused by an older version of anomalize installed from Github. Reinstallation of the latest version from CRAN (along with dependencies) fixed the problem. — neilfws, Jul 25 '18 at 12:08

score 1 · Answer 1 · answered Apr 24 '18 at 21:27

This is a stumper. I worked through that tutorial a few weeks ago, so I'm comparing your data to the tidyverse_cran_downloads and I don't see what the difference is. It might not like date as class POSIX at any time, but I really don't see how that could matter. I also ran through it with made up data in your same layout and it worked. Hopefully this helps in some way.

counts <- data.frame(date = as.Date(c("2013-09-13", "2014-01-23", "2014-06-24", "2017-04-25", "2017-04-26", "2017-06-28", "2017-09-26")), word = c("hr", "hr", "hr", "hr", "hr", "hr", "hr"), count = c(2, 6, 2, 3, 2, 4, 4), stringsAsFactors = FALSE)

counts <- counts %>% 
tibbletime::as_tbl_time(index = date)

counts %>% 
time_decompose(count, method = "stl") %>% 
anomalize(remainder) %>% 
time_recompose() %>% 
plot_anomalies(time_recomposed = TRUE, ncol = 3, alpha_dots = 0.5)

score 0 · Answer 2 · edited Aug 31 '18 at 00:03

0

I had similar problem and in my case group_by before time_decompose helped:

y %>% group_by(word) %>%
    time_decompose(count,method='stl')%>%
    anomalize(remainder,method = 'iqr') %>%
    time_recompose()%>% plot_anomalies(time_recomposed = T,ncol=5,alpha_dots = 0.25)

edited Aug 31 '18 at 00:03

Stephen Rauch

47,830
31
106
135

answered Aug 30 '18 at 23:31

wojciechowa

1

score 0 · Answer 3 · answered Jul 18 '22 at 14:44

0

Here are the two steps which helped in my case. Think the latter one is more important.

Rename date column to index.
Removing older version of 'rlang' and updating it.

answered Jul 18 '22 at 14:44

Thomas Halamuda

21
5

Anomalize error: no date or datetime column found

3 Answers3