Skip to contents

Add Random Noise to Measurements

Source: R/augmentation.R
step_measure_augment_noise.Rd

step_measure_augment_noise() creates a specification of a recipe step that adds controlled random noise to spectral data for data augmentation.

Usage

step_measure_augment_noise(
  recipe,
  sd = 0.01,
  distribution = c("gaussian", "uniform"),
  relative = TRUE,
  measures = NULL,
  role = NA,
  trained = FALSE,
  skip = TRUE,
  id = recipes::rand_id("measure_augment_noise")
)

Arguments

recipe

A recipe object.

sd

Standard deviation of noise. If relative = TRUE (default), this is relative to the signal range (0.01 = 1% of range). If relative = FALSE, this is the absolute noise level.

distribution

Noise distribution: "gaussian" (default) or "uniform".

relative

Logical. If TRUE (default), sd is relative to signal range.

measures

An optional character vector of measure column names.

role

Not used.

trained

Logical indicating if the step has been trained.

skip

Logical. Should the step be skipped when baking? Default is TRUE, meaning augmentation only applies during training.

id

Unique step identifier.

Value

An updated recipe with the new step added.

Details

Data augmentation adds variability to training data to improve model robustness. Adding noise simulates measurement uncertainty and helps models generalize better.

Default behavior (skip = TRUE): The augmentation is only applied during prep() on training data. When bake() is called on new data, the step is skipped.

Reproducibility: The noise is deterministic based on the row content, so the same input always produces the same augmented output within a session.

See also

Other measure-augmentation: step_measure_augment_scale(), step_measure_augment_shift()

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_augment_noise(sd = 0.02) |>
  prep()

# Noise only applied to training data
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