CEEMDAN的使用方法

我只是help文档的搬运工。

Help on class CEEMDAN in module PyEMD.CEEMDAN:

class CEEMDAN(builtins.object)
   CEEMDAN(trials: int = 100, epsilon: float = 0.005, ext_EMD=None, parallel: bool = False, **kwargs)

使用方法举例：
 |  Example:
 |  
 |  >>> ceemdan = CEEMDAN(range_thr=0.001, total_power_thr=0.01)
 |  
 |  To perform the decomposition one can either use directly initiated object,
 |  or use the `ceemdan` method. The following two lines produce the same output:
 |  
 |  >>> ceemdan = CEEMDAN()
 |  >>> c_imfs = ceemdan(signal)
 |  >>> c_imfs = ceemdan.ceemdan(signal)
 |  
参数含义： 
 |  Parameters
 |  ----------
 |  
 |  trials : int (default: 100)
 |      Number of trials or EMD performance with added noise.
 |  epsilon : float (default: 0.005)
 |      Scale for added noise (:math:`\epsilon`) which multiply std :math:`\sigma`:
 |      :math:`\beta = \epsilon \cdot \sigma`
 |  ext_EMD : EMD (default: None)
 |      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.
 |  parallel : bool (default: False)
 |      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.
 |  processes : int or None (optional)
 |      Number of processes harness when executing in parallel mode.
 |      The value should be between 1 and max that depends on your hardware.
 |  noise_scale : float (default: 1)
 |      Scale (amplitude) of the added noise.
 |  noise_kind : str (default: "normal")
 |      What type of noise to add. Allowed are "normal" (default) and "uniform".
 |  range_thr : float (default: 0.01)
 |      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.
 |  total_power_thr : float (default: 0.05)
 |      Signal's power threshold. Finishes decomposition if sum(abs(r)) < thr.
 |  beta_progress : bool (default: True)
 |      Flag whether to scale all noise IMFs by their 1st IMF's standard deviation.