Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Error while trying to create recipe #149

Open
beschlemper opened this issue Jul 26, 2023 · 0 comments
Open

Error while trying to create recipe #149

beschlemper opened this issue Jul 26, 2023 · 0 comments

Comments

@beschlemper
Copy link

beschlemper commented Jul 26, 2023

Hello,

I have a simple database with a column of dates and other columns with the variables. I had no problem with ARIMA or Prophet, but I got this error when trying to create the recipe:

`recipe <- recipe(value ~ Ano, training(splits)) %>%

  • step_timeseries_signature(Ano) %>%
  • step_fourier(Ano, period = 1, K = 3) %>%
  • step_dummy(all_nominal())

recipe %>% prep() %>% juice()
A tibble: 198 × 50
Ano value Ano_index.num Ano_year Ano_year.iso Ano_half Ano_quarter Ano_month

1 0001-07-01 856768 -62119958400 1 1 2 3 7
2 0002-07-01 755395 -62088422400 2 2 2 3 7
3 0003-07-01 814879 -62056886400 3 3 2 3 7
4 0004-07-01 708760 -62025264000 4 4 2 3 7
5 0005-07-01 707114 -61993728000 5 5 2 3 7
6 0006-07-01 727197 -61962192000 6 6 2 3 7
7 0007-07-01 802187 -61930656000 7 7 2 3 7
8 0008-07-01 764448 -61899033600 8 8 2 3 7
9 0009-07-01 777312 -61867497600 9 9 2 3 7
10 0010-07-01 797514 -61835961600 10 10 2 3 7
ℹ 188 more rows
ℹ 42 more variables: Ano_month.xts , Ano_day , Ano_hour , Ano_minute ,
Ano_second , Ano_hour12 , Ano_am.pm , Ano_wday , Ano_wday.xts ,
Ano_mday , Ano_qday , Ano_yday , Ano_mweek , Ano_week ,
Ano_week.iso , Ano_week2 , Ano_week3 , Ano_week4 , Ano_mday7 ,
Ano_sin1_K1 , Ano_cos1_K1 , Ano_sin1_K2 , Ano_cos1_K2 ,
Ano_sin1_K3 , Ano_cos1_K3 , Ano_month.lbl_01 , Ano_month.lbl_02 , …
ℹ Use print(n = ...) to see more rows
Warning message:
There were 6 warnings in dplyr::mutate().
The first warning was:
ℹ In argument: Ano_sin1_K1 = timetk::fourier_vec(...).
Caused by warning in as.POSIXct(x) %>% as.numeric() %>% as.integer():
! NAs introduced by coercion to integer range
ℹ Run dplyr::last_dplyr_warnings() to see the 5 remaining warnings. `

The first rows of the database is just like this but it has 210 rows:

ano var1 var2 var3 var4 var5 var6 var7 var8 var9
1 856768 0 213561 0 0 36874 0 0 150408
2 755395 407,691 198381 0 0 38085 0 0 131574
3 814879 353,735 225410 0 0 25944 0 0 178831
4 708760 340,257 211353 1948448 177159,6 29263 2378163 789316,4 141362
5 707114 332,53 196299 1820596 165823,5 46797 2131468 719312,8 187400
6 727197 375,972 194509 2063582 186949,7 37183 2571621 857821,8 191155

I already used the same script for other database, but with this one I keep getting this error.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant