tests/testthat/_snaps/time-series-dataset.md

In mlverse/tft: Implementation of Temporal Fusion Transformer

can create a spec

Code
  print(spec)
Output
  A <tft_dataset_spec> with:

  x `lookback` and `horizon` are not set. Use `spec_time_splits()`

  -- Covariates: 
  x `index` is not set. Use `spec_covariate_index()` to set it.
  x `keys` is not set. Use `spec_covariate_keys()` to set it.
  ! `static` is not set. Use `spec_covariate_static()` to set it.
  ! `known` is not set. Use `spec_covariate_known()` to set it.
  ! `unknown` is not set. Covariates that are not listed as other types are considered `unknown`.

  i Call `prep()` to prepare the specification.

A <tft_dataset_spec> with:

v lookback = 52 and horizon = 4.

-- Covariates: 
v `index`: Date
v `keys`: <list: Store, Dept>
v `static`: <list: Type, Size>
v `known`: <list: starts_with("MarkDown"), starts_with("Date_"), starts_with("na_ind")>
! `unknown` is not set. Covariates that are not listed as other types are considered `unknown`.

i Call `prep()` to prepare the specification.

A <prepared_tft_dataset_spec> with:

v lookback = 52 and horizon = 4.
v The number of possible slices is 320

-- Covariates: 
v `index`: Date
v `keys`: Store and Dept
v `static`: Type and Size
v `known`: MarkDown1, MarkDown2, MarkDown3, MarkDown4, MarkDown5, na_ind_MarkDown1, na_ind_MarkDown2, na_ind_MarkDown3, na_ind_MarkDown4, and na_ind_MarkDown5
v `unknown`: IsHoliday, Temperature, Fuel_Price, CPI, Unemployment, and intercept
i Variables that are not specified in other types are considered `unknown`.

i Call `transform()` to apply this spec to a different dataset.

mlverse/tft documentation built on June 19, 2022, 4:31 a.m.