Skip to contents

Orthogonal Signal Correction (OSC)

Source: R/scatter-extended.R
step_measure_osc.Rd

step_measure_osc() creates a specification of a recipe step that applies Orthogonal Signal Correction to remove variation orthogonal to the outcome.

Usage

step_measure_osc(
  recipe,
  n_components = 1L,
  tolerance = 1e-06,
  max_iter = 100L,
  measures = NULL,
  role = NA,
  trained = FALSE,
  weights = NULL,
  loadings = NULL,
  skip = FALSE,
  id = recipes::rand_id("measure_osc")
)

Arguments

recipe

A recipe object.

n_components

Number of orthogonal components to remove. Default is 1.

tolerance

Convergence tolerance for NIPALS algorithm. Default is 1e-6.

max_iter

Maximum iterations for NIPALS. Default is 100.

measures

An optional character vector of measure column names.

role

Not used.

trained

Logical indicating if the step has been trained.

weights

The learned orthogonal weights (after training).

loadings

The learned orthogonal loadings (after training).

skip

Logical. Should the step be skipped when baking?

id

Unique step identifier.

Value

An updated recipe with the new step added.

Details

Orthogonal Signal Correction (OSC) removes variation in X that is orthogonal to Y (the outcome). This is useful for removing systematic variation that is not related to the response.

Algorithm:

  1. Compute initial score t from Y using SVD

  2. Orthogonalize t with respect to Y

  3. Iterate NIPALS to find orthogonal components

  4. Remove orthogonal components from X

Important:

  • The recipe must have at least one outcome variable with role "outcome"

  • Outcomes are automatically detected from the recipe's role definitions

  • Multiple outcomes are supported (multivariate Y)

OSC was originally described by Wold et al. (1998) for NIR spectroscopy.

References

Wold, S., Antti, H., Lindgren, F., and Ohman, J. (1998). Orthogonal signal correction of near-infrared spectra. Chemometrics and Intelligent Laboratory Systems, 44(1-2), 175-185.

See also

Other measure-preprocessing: step_measure_absorbance(), step_measure_calibrate_x(), step_measure_calibrate_y(), step_measure_derivative(), step_measure_derivative_gap(), step_measure_emsc(), step_measure_kubelka_munk(), step_measure_log(), step_measure_map(), step_measure_msc(), step_measure_normalize_istd(), step_measure_ratio_reference(), step_measure_snv(), step_measure_subtract_blank(), step_measure_subtract_reference(), step_measure_transmittance()

Examples

library(recipes)

rec <- recipe(water + fat + protein ~ ., data = meats_long) |>
  update_role(id, new_role = "id") |>
  step_measure_input_long(transmittance, location = vars(channel)) |>
  step_measure_osc(n_components = 2) |>
  prep()

bake(rec, new_data = NULL)
#> # A tibble: 215 × 5
#>       id water   fat protein .measures
#>    <int> <dbl> <dbl>   <dbl>    <meas>
#>  1     1  60.5  22.5    16.7 [100 × 2]
#>  2     2  46    40.1    13.5 [100 × 2]
#>  3     3  71     8.4    20.5 [100 × 2]
#>  4     4  72.8   5.9    20.7 [100 × 2]
#>  5     5  58.3  25.5    15.5 [100 × 2]
#>  6     6  44    42.7    13.7 [100 × 2]
#>  7     7  44    42.7    13.7 [100 × 2]
#>  8     8  69.3  10.6    19.3 [100 × 2]
#>  9     9  61.4  19.9    17.7 [100 × 2]
#> 10    10  61.4  19.9    17.7 [100 × 2]
#> # ℹ 205 more rows