tests/testthat/_snaps/forge-formula.md

In tidymodels/hardhat: Construct Modeling Packages

novel predictor levels are caught

Code
  xx1 <- forge(new, x1$blueprint)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

Code
  xx2 <- forge(new, x2$blueprint)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

novel predictor levels can be ignored

Code
  xx1 <- forge(new, x1$blueprint)

Code
  xx2 <- forge(new, x2$blueprint)

novel levels are handled correctly when the new column is a character

Code
  xx1 <- forge(new, x1$blueprint)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

Code
  xx2 <- forge(new, x2$blueprint)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

novel levels are ignored correctly when the new column is a character

Code
  xx1 <- forge(new, x1$blueprint)

Code
  xx2 <- forge(new, x2$blueprint)

novel outcome levels are caught

Code
  xx1 <- forge(new, x1$blueprint, outcomes = TRUE)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

Code
  xx2 <- forge(new, x2$blueprint, outcomes = TRUE)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

novel outcome levels are always caught, even if allow_novel_levels = TRUE

Code
  xx1 <- forge(new, x1$blueprint, outcomes = TRUE)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

Code
  xx2 <- forge(new, x2$blueprint, outcomes = TRUE)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

missing predictor levels are restored silently

Code
  xx1 <- forge(new, x1$blueprint)

Code
  xx2 <- forge(new, x2$blueprint)

Code
  yy1 <- forge(new2, x1$blueprint)

Code
  yy2 <- forge(new2, x2$blueprint)

can be both missing levels and have new levels

Code
  mold(y ~ f, dat, blueprint = bp2)
Condition
  Error in `recompose()`:
  ! `data` must only contain numeric columns.
  i These columns aren't numeric: "f".

Code
  xx <- forge(new, x1$blueprint)
Condition
  Warning:
  Novel levels found in column 'f': 'e'. The levels have been removed, and values have been coerced to 'NA'.

can be both missing levels and have new levels that get ignored

Code
  xx <- forge(new, x$blueprint)

NA factor data never triggers a novel level warning (#131)

Code
  xx <- forge(new, x$blueprint)

allow_novel_levels works right with character predictors

Code
  out <- forge(df2, x$blueprint)
Condition
  Warning:
  Novel levels found in column 'x': 'd'. The levels have been removed, and values have been coerced to 'NA'.

tidymodels/hardhat documentation built on Sept. 26, 2024, 11:38 p.m.