Error: Error time_decompose(): when perform Anomaly Detection in R

user_4529548 0
Viewed 415

Here mydata

zem=zem = read.table(
V1
75
74,7
74,4
74,1
73,8
75,5
73,3
73,1
72,9
73
72,8
72,3
72,1
71,9
71,7
71,6
71,3
71,4
71,3
71,2
71,1
70
69,5
69
68,5)

I want detect anomaly value. So i decided use library(anomalize). Code below

library(anomalize) #tidy anomaly detectiom
library(tidyverse) #tidyverse packages like dplyr, ggplot, tidyr 



  zem %>% 
      time_decompose(V1) %>%
      anomalize(remainder) %>%
      time_recompose() %>%
      filter(anomaly == 'Yes')

an i get the error

Error: Error time_decompose(): Object is not of class tbl_df or tbl_time.

What's wrong? How can i get desired result?

     V1 Anomaly
1  75.0      no
2  74.7      no
3  74.4      no
4  74.1      no
5  73.8      no
6  75.5     yes
7  73.3      no
8  73.1      no
9  72.9      no
10 73.0      no
11 72.8      no
12 72.3      no
13 72.1      no
14 71.9      no
15 71.7      no
16 71.6      no
17 71.3      no
18 71.4      no
19 71.3      no
20 71.2      no
21 71.1      no
22 70.0      no
23 69.5      no
24 69.0      no
25 68.5      no

i just tried modify this code for my task

https://towardsdatascience.com/tidy-anomaly-detection-using-r-82a0c776d523
tidyverse r
1 Answers

The time_decompose() function requires data in the form of:

A tibble or tbl_time object

(from ?time_decompose)

Perhaps zem is a data.frame? You can include as_tibble() in the pipe to make sure it is a tibble ahead of time.

In addition, it expects to work on time based data:

It is designed to work with time-based data, and as such must have a column that contains date or datetime information.

I added to your test data a column with dates. Here is a working example:

library(anomalize) 
library(tidyverse)  

zem$date <- as.Date(Sys.Date() + 1:nrow(zem))

zem %>% 
  as_tibble() %>%
  time_decompose(V1) %>%
  anomalize(remainder) %>%
  time_recompose() %>%
  filter(anomaly == 'Yes')

Output

Converting from tbl_df to tbl_time.
Auto-index message: index = date
frequency = 7 days
trend = 12.5 days
# A time tibble: 4 x 10
# Index: date
  date       observed season trend remainder remainder_l1 remainder_l2 anomaly recomposed_l1 recomposed_l2
  <date>        <dbl>  <dbl> <dbl>     <dbl>        <dbl>        <dbl> <chr>           <dbl>         <dbl>
1 2020-12-28     75.5  0.782  73.8     0.934       -0.555        0.624 Yes              74.0          75.2
2 2021-01-04     72.1  0.782  72.3    -0.996       -0.555        0.624 Yes              72.5          73.7
3 2021-01-10     71.3 -0.229  70.7     0.789       -0.555        0.624 Yes              70.0          71.1
4 2021-01-12     71.1 -0.220  70.1     1.24        -0.555        0.624 Yes              69.3          70.5

Here is a visual of anomalies detected:

zem %>% 
  as_tibble() %>%
  time_decompose(V1) %>%
  anomalize(remainder) %>%
  plot_anomaly_decomposition()

Plot

anomaly plot

Data

zem <- structure(list(V1 = c(75, 74.7, 74.4, 74.1, 73.8, 75.5, 73.3, 
73.1, 72.9, 73, 72.8, 72.3, 72.1, 71.9, 71.7, 71.6, 71.3, 71.4, 
71.3, 71.2, 71.1, 70, 69.5, 69, 68.5), date = structure(c(18619, 
18620, 18621, 18622, 18623, 18624, 18625, 18626, 18627, 18628, 
18629, 18630, 18631, 18632, 18633, 18634, 18635, 18636, 18637, 
18638, 18639, 18640, 18641, 18642, 18643), class = "Date")), row.names = c(NA, 
-25L), class = "data.frame")
user_34606700
Related