NaiveNormalize#

class torch_ecg._preprocessors.NaiveNormalize(mean: float | int | ndarray[tuple[Any, ...], dtype[_ScalarT]] = 0.0, std: float | int | ndarray[tuple[Any, ...], dtype[_ScalarT]] = 1.0, per_channel: bool = False, **kwargs: Any)[source]#

Bases: Normalize

Naive normalization.

Naive normalization defined as

\[\frac{sig - m}{s}\]
Parameters:

  • mean (float or int or numpy.ndarray, default 0.0) – Value(s) to be subtracted.

  • std (float or int or numpy.ndarray, default 1.0) – Value(s) to be divided.

  • per_channel (bool, default False) – If True, normalization will be done per channel.

Examples

from torch_ecg.cfg import DEFAULTS
sig = DEFAULTS.RNG.randn(1000)
pp = NaiveNormalize()
sig, _ = pp(sig, 500)
extra_repr_keys() → List[str][source]#

Extra keys for __repr__() and __str__().