amep.utils.detect2peaks

amep.utils.detect2peaks(xdata: ndarray, ydata: ndarray, nav: int = 11, distance: int = 8, height: float = 0.1, width: int = 4) → tuple[list, list, ndarray, ndarray]

Detects the two highest peaks in the given data after smoothing it with a running average. Includes also global maxima at the boundaries.

Notes

This method only produces correct results if there are at most two

peaks present! Please check the behavior in your individual case.

Parameters:

• xdata (np.ndarray) – x values as 1D array of floats.

• ydata (np.ndarray) – y values as 1D array of floats (same shape as xdata).

• nav (int, optional) – Number of points to average over via running mean. The default is 11.

• distance (int, optional) – Minimum distance in samples between two peaks. The default is 8.

• height (float, optional) – Minium height of a peak. The default is 0.1.

• width (int, optional) – Minimum peak width in samples. The default is 4.

Returns:

• low (list) – Coordinates [x,y] of the peak at smaller x value.

• high (list) – Coordinates [x,y] of the peak at larger x value.

• avydata (np.ndarray) – Averaged y values.

• avxdata (np.ndarray) – Averaged x values (same shape as avydata).

Examples

>>> import amep
>>> import numpy as np
>>> x = np.linspace(1,100,1000)
>>> y = np.exp(-(x-20)**2/5)
>>> y += np.exp(-(x-70)**2/10)
>>> y += np.random.rand(len(x))*0.1
>>> low, high, avy, avx = amep.utils.detect2peaks(
...     x, y, nav=10, width=2
... )
>>> fig, axs = amep.plot.new()
>>> axs.plot(x, y, label='original', c='k', ls='')
>>> axs.plot(avx, avy, label='averaged', c='r', marker='')
>>> axs.plot(
...     low[0], low[1], linestyle='', marker='.',
...     c='b', markersize=10, label='peaks'
... )
>>> axs.plot(
...     high[0], high[1], linestyle='',
...     marker='.', c='b', markersize=10
... )
>>> axs.legend()
>>> axs.set_xlabel(r'$x$')
>>> axs.set_ylabel(r'$f(x)$')
>>> fig.savefig('./figures/utils/utils-detect2peaks.png')
>>>