Type: Package
Title: Re-Scale Vectors and Time-Series Features
Version: 0.1.2
Date: 2024-02-28
Maintainer: Trent Henderson <then6675@uni.sydney.edu.au>
Description: Provides standardized access to a range of re-scaling methods for numerical vectors and time-series features calculated within the 'theft' ecosystem.
BugReports: https://github.com/hendersontrent/normaliseR/issues
License: MIT + file LICENSE
Encoding: UTF-8
Depends: R (≥ 3.5.0)
Imports: rlang, stats, dplyr, scales
Suggests: knitr, markdown, rmarkdown, pkgdown, testthat (≥ 3.0.0)
RoxygenNote: 7.2.2
VignetteBuilder: knitr
Config/testthat/edition: 3
URL: https://hendersontrent.github.io/normaliseR/
NeedsCompilation: no
Packaged: 2024-02-28 00:25:29 UTC; trenthenderson
Author: Trent Henderson [cre, aut]
Repository: CRAN
Date/Publication: 2024-02-29 11:50:02 UTC

Re-Scale Vectors and Time-Series Features

Description

Re-scale Vectors and Time-Series Features

Rescales a numeric vector using maximum absolute scaling

Description

z_{i} = \frac{x_{i}}{\text{max}(\mathbf{x})}

Usage

maxabs_scaler(x)

Arguments

x

numeric vector

Value

numeric vector

Author(s)

Trent Henderson

Rescales a numeric vector into the unit interval [0,1]

Description

z_{i} = \frac{x_{i} - \text{min}(\mathbf{x})}{\text{max}(\mathbf{x}) - \text{min}(\mathbf{x})}

Usage

minmax_scaler(x)

Arguments

x

numeric vector

Value

numeric vector

Author(s)

Trent Henderson

Scale each feature vector into a user-specified range for visualisation and modelling

Description

'normalise()' and 'normalize()' are synonyms.

Usage

normalise(
  data,
  norm_method = c("zScore", "Sigmoid", "RobustSigmoid", "MinMax", "MaxAbs"),
  unit_int = FALSE
)

normalize(
  data,
  norm_method = c("zScore", "Sigmoid", "RobustSigmoid", "MinMax", "MaxAbs"),
  unit_int = FALSE
)

Arguments

data

either a feature_calculations object containing the raw feature matrix produced by theft::calculate_features or a vector of class numeric containing values to be rescaled

norm_method

character denoting the rescaling/normalising method to apply. Can be one of "zScore", "Sigmoid", "RobustSigmoid", "MinMax", or "MaxAbs". Defaults to "zScore"

unit_int

Boolean whether to rescale into unit interval [0,1] after applying normalisation method. Defaults to FALSE

Value

either an object of class feature_calculations object or a numeric vector depending on the data type supplied to data

Author(s)

Trent Henderson

Rescales a numeric vector using an outlier-robust Sigmoidal transformation

Description

z_{i} = \left[1 + \exp\left(-\frac{x_{i} - \text{median}(\mathbf{x})}{\text{IQR}(\mathbf{x})/{1.35}}\right)\right]^{-1}

Usage

robustsigmoid_scaler(x)

Arguments

x

numeric vector

Value

numeric vector

Author(s)

Trent Henderson

References

Fulcher, Ben D., Little, Max A., and Jones, Nick S. Highly Comparative Time-Series Analysis: The Empirical Structure of Time Series and Their Methods. Journal of The Royal Society Interface 10(83), (2013).

Rescales a numeric vector using a Sigmoidal transformation

Description

z_{i} = \left[1 + \exp(-\frac{x_{i} - \mu}{\sigma})\right]^{-1}

Usage

sigmoid_scaler(x)

Arguments

x

numeric vector

Value

numeric vector

Author(s)

Trent Henderson

Rescales a numeric vector into z-scores

Description

z_{i} = \frac{x_{i} - \mu}{\sigma}

Usage

zscore_scaler(x)

Arguments

x

numeric vector

Value

numeric vector

Author(s)

Trent Henderson