from typing import Any, List
from PyEMD import CEEMDAN, EMD # type: ignore
from ..signal import Signal
[docs]def get_IMFs_ceemdan(
data: Signal,
number_seedman=30,
epsilon=0.005,
ext_EMD: Any = None,
parallel=False,
processes=None,
noise_scale=1,
noise_kind="normal",
range_thr=0.01,
total_power_thr=0.05,
) -> List[Signal]:
'''Empirical mode decomposition (EMD).
Using CEEMDAN from PyEMD (https://pyemd.readthedocs.io/) to calculate IMFs
and return them as a list of `Signals`.
About empirical mode decomposition on
https://en.wikipedia.org/wiki/Hilbert%E2%80%93Huang_transform#Techniques
Args:
data (Signal): _description_
number_seedman (int, optional): _description_. Defaults to 30.
epsilon (float, optional): Scale for added noise (\\epsilon) which
multiply std \\sigma: \beta = \\epsilon \\cdot \\sigma.
Defaults to 0.005.
ext_EMD (Any, optional): One can pass EMD object defined outside,
which will be used to compute IMF decompositions in each trial.
If none is passed then EMD with default options is used.
Defaults to None.
parallel (bool, optional): Flag whether to use multiprocessing in
EEMD execution. Since each EMD(s+noise) is independent this should
improve execution speed considerably. *Note* that it's disabled by
default because it's the most common problem when CEEMDAN takes too
long time to finish. If you set the flag to True, make also sure to
set processes to some reasonable value. Defaults to False.
processes (_type_, optional): Number of processes harness when
executing in parallel mode. The value should be between 1 and
max that depends on your hardware. Defaults to None.
noise_scale (int, optional): Scale (amplitude) of the added noise.
Defaults to 1.
noise_kind (str, optional): What type of noise to add. Allowed
are "normal" (default) and "uniform". Defaults to "normal".
range_thr (float, optional): Range threshold used as an IMF check.
The value is in percentage compared to initial signal's amplitude.
If absolute amplitude (max - min) is below the range_thr then the
decomposition is finished. Defaults to 0.01.
total_power_thr (float, optional): Signal's power threshold.
Finishes decomposition if sum(abs(r)) < thr. Defaults to 0.05.
Returns:
List[Signal]: List of Signals expected IMFs.
'''
y = data.y
emd = CEEMDAN(
number_seedman,
epsilon,
ext_EMD,
parallel,
processes=processes,
noise_scale=noise_scale,
noise_kind=noise_kind,
range_thr=range_thr,
total_power_thr=total_power_thr,
)
IMFs = emd(y)
result = [Signal(data.x.copy(), k) for k in IMFs]
return result
[docs]def get_IMFs_emd(data: Signal) -> List[Signal]:
'''"Empirical mode decomposition (EMD).
Using EMD from PyEMD (https://pyemd.readthedocs.io/) to calculate IMFs
and return them as a list of `Signals`.
About empirical mode decomposition on
https://en.wikipedia.org/wiki/Hilbert%E2%80%93Huang_transform#Techniques
Args:
data (Signal): signal to calculate IMFs.
Returns:
List[Signal]: List of Signals expected IMFs.
'''
y = data.y
emd = EMD()
IMFs = emd(y)
result = [Signal(data.x.copy(), k) for k in IMFs]
return result