Skip to contents

Log Transformation

Source: R/spectral-math.R
step_measure_log.Rd

step_measure_log() creates a specification of a recipe step that applies a logarithmic transformation to measurement values.

Usage

step_measure_log(
  recipe,
  base = exp(1),
  offset = 0,
  measures = NULL,
  role = NA,
  trained = FALSE,
  skip = FALSE,
  id = recipes::rand_id("measure_log")
)

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.

base

The base of the logarithm. Default is exp(1) (natural log). Use 10 for log10 transformation.

offset

A numeric offset added to values before taking the log. Default is 0. Use a small positive value (e.g., 1 for log1p) to handle zero or near-zero values.

measures

An optional character vector of measure column names to process. If NULL (the default), all measure columns will be processed.

role

Not used by this step since no new variables are created.

trained

A logical to indicate if the step has been trained.

skip

A logical. Should the step be skipped when baking?

id

A character string that is unique to this step.

Value

An updated version of recipe with the new step added.

Details

This step applies the transformation:

$$y' = \log_b(y + \text{offset})$$

where \(b\) is the base.

Log transformation is commonly used for:

  • Variance stabilization

  • Normalizing skewed distributions

  • Converting multiplicative relationships to additive

Warning: Non-positive values (after offset) will produce -Inf or NaN.

The measurement locations are preserved unchanged.

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_map(), step_measure_msc(), step_measure_normalize_istd(), step_measure_osc(), step_measure_ratio_reference(), step_measure_snv(), step_measure_subtract_blank(), step_measure_subtract_reference(), step_measure_transmittance()

Examples

library(recipes)

# Natural log transformation
rec <- recipe(water + fat + protein ~ ., data = meats_long) |>
  update_role(id, new_role = "id") |>
  step_measure_input_long(transmittance, location = vars(channel)) |>
  step_measure_log(offset = 1) |>
  prep()

# Log10 transformation
rec2 <- recipe(water + fat + protein ~ ., data = meats_long) |>
  update_role(id, new_role = "id") |>
  step_measure_input_long(transmittance, location = vars(channel)) |>
  step_measure_log(base = 10) |>
  prep()