Skip to contents

Spectral Binning

Source: R/feature-engineering.R
step_measure_bin.Rd

step_measure_bin() creates a specification of a recipe step that reduces a spectrum to fewer points by averaging within bins.

Usage

step_measure_bin(
  recipe,
  n_bins = NULL,
  bin_width = NULL,
  method = c("mean", "sum", "median", "max"),
  measures = NULL,
  role = NA,
  trained = FALSE,
  bin_breaks = NULL,
  skip = FALSE,
  id = recipes::rand_id("measure_bin")
)

Arguments

recipe

A recipe object.

n_bins

Number of bins (mutually exclusive with bin_width).

bin_width

Width of each bin in location units (mutually exclusive with n_bins).

method

Aggregation method: "mean" (default), "sum", "median", or "max".

measures

An optional character vector of measure column names.

role

Not used (modifies existing data).

trained

Logical indicating if the step has been trained.

bin_breaks

The computed bin breaks (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

This step reduces the number of points in each spectrum by dividing the x-axis into bins and aggregating values within each bin. The result replaces the .measures column with the binned data.

This is useful for:

  • Reducing data dimensionality

  • Decreasing noise through averaging

  • Speeding up downstream processing

  • Aligning data from different resolutions

The bin boundaries are determined during prep() from the training data and stored for consistent application to new data.

See also

Other measure-features: step_measure_integrals(), step_measure_moments(), step_measure_ratios()

Examples

library(recipes)

# Bin to 20 points
rec <- recipe(water + fat + protein ~ ., data = meats_long) |>
  update_role(id, new_role = "id") |>
  step_measure_input_long(transmittance, location = vars(channel)) |>
  step_measure_bin(n_bins = 20) |>
  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  [20 × 2]
#>  2     2  46    40.1    13.5  [20 × 2]
#>  3     3  71     8.4    20.5  [20 × 2]
#>  4     4  72.8   5.9    20.7  [20 × 2]
#>  5     5  58.3  25.5    15.5  [20 × 2]
#>  6     6  44    42.7    13.7  [20 × 2]
#>  7     7  44    42.7    13.7  [20 × 2]
#>  8     8  69.3  10.6    19.3  [20 × 2]
#>  9     9  61.4  19.9    17.7  [20 × 2]
#> 10    10  61.4  19.9    17.7  [20 × 2]
#> # ℹ 205 more rows