wildboar.datasets.preprocess

Utilities for preprocessing time series.

Module Contents

Functions

maxabs_scale(x)	Scale each time series by its maximum absolute value.
minmax_scale(x[, min, max])	Scale x along the time dimension.
named_preprocess(name)	Get a named preprocessor.
standardize(x)	Scale x along the time dimension.
truncate(x[, n_shortest])	Truncate x to the shortest sequence.

wildboar.datasets.preprocess.maxabs_scale(x)

Scale each time series by its maximum absolute value.

Parameters:

xndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The samples.

Returns:

ndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The transformed samples.

wildboar.datasets.preprocess.minmax_scale(x, min=0, max=1)

Scale x along the time dimension.

Each time series is scaled such that each value is between min and max.

Parameters:

xndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The samples.

minfloat, optional

The minimum value.

maxfloat, optional

The maximum value.

Returns:

ndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The transformed samples.

wildboar.datasets.preprocess.named_preprocess(name)

Get a named preprocessor.

Parameters:

namestr

The name of the preprocessor.

Returns:

callable

The preprocessor function.

wildboar.datasets.preprocess.standardize(x)

Scale x along the time dimension.

The resulting array will have zero mean and unit standard deviation.

Parameters:

xndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The samples.

Returns:

ndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The standardized samples.

wildboar.datasets.preprocess.truncate(x, n_shortest=None)

Truncate x to the shortest sequence.

Parameters:

xndarray of shape (n_samples, n_timestep) or (n_samples, n_dims, n_timestep)

The samples.

n_shortestint, optional

The maximum size.

Returns:

ndarray of shape (n_samples, n_shortest) or (n_samples, n_dims, n_shortest)

The truncated samples.