In almost all cases, use step_dct instead. The discrete cosine transform has better compaction in general. DCT coefficients are always real, while DFT returns complex numbers, which increases the number of dimensions.
step_fft( recipe, ..., role = "predictor", trained = FALSE, k = 4, dct = TRUE, series = NULL, skip = FALSE, id = recipes::rand_id("tff") )
A recipe object. The step will be added to the sequence of operations for this recipe.
One or more selector functions to choose which
variables are affected by the step. See
For model terms created by this step, what analysis role should they be assigned?. By default, the function assumes that the new columns created from the original variables will be used as predictors in a model.
A logical to indicate if the quantities for preprocessing have been estimated.
The number of discrete Fourier transform coefficients
A logical. Should the step be skipped when the
recipe is baked by
A character string that is unique to this step to identify it.
A list of the length and coefficient indices for each time series passed to the step, created once the step has been trained.
Sayood, K. Introduction to Data Compression