Return (maximum - minimum) along the given dimension (i.e. peak-to-peak value). .. warning:: `ptp` preserves the data type of the array. This means the return value for an input of signed integers with n bits (e.g. `np.int8`, `np.int16`,
(self, axis=None, out=None, fill_value=None, keepdims=False)
| 6077 | return out |
| 6078 | |
| 6079 | def ptp(self, axis=None, out=None, fill_value=None, keepdims=False): |
| 6080 | """ |
| 6081 | Return (maximum - minimum) along the given dimension |
| 6082 | (i.e. peak-to-peak value). |
| 6083 | |
| 6084 | .. warning:: |
| 6085 | `ptp` preserves the data type of the array. This means the |
| 6086 | return value for an input of signed integers with n bits |
| 6087 | (e.g. `np.int8`, `np.int16`, etc) is also a signed integer |
| 6088 | with n bits. In that case, peak-to-peak values greater than |
| 6089 | ``2**(n-1)-1`` will be returned as negative values. An example |
| 6090 | with a work-around is shown below. |
| 6091 | |
| 6092 | Parameters |
| 6093 | ---------- |
| 6094 | axis : {None, int}, optional |
| 6095 | Axis along which to find the peaks. If None (default) the |
| 6096 | flattened array is used. |
| 6097 | out : {None, array_like}, optional |
| 6098 | Alternative output array in which to place the result. It must |
| 6099 | have the same shape and buffer length as the expected output |
| 6100 | but the type will be cast if necessary. |
| 6101 | fill_value : scalar or None, optional |
| 6102 | Value used to fill in the masked values. |
| 6103 | keepdims : bool, optional |
| 6104 | If this is set to True, the axes which are reduced are left |
| 6105 | in the result as dimensions with size one. With this option, |
| 6106 | the result will broadcast correctly against the array. |
| 6107 | |
| 6108 | Returns |
| 6109 | ------- |
| 6110 | ptp : ndarray. |
| 6111 | A new array holding the result, unless ``out`` was |
| 6112 | specified, in which case a reference to ``out`` is returned. |
| 6113 | |
| 6114 | Examples |
| 6115 | -------- |
| 6116 | >>> import numpy as np |
| 6117 | >>> x = np.ma.MaskedArray([[4, 9, 2, 10], |
| 6118 | ... [6, 9, 7, 12]]) |
| 6119 | |
| 6120 | >>> x.ptp(axis=1) |
| 6121 | masked_array(data=[8, 6], |
| 6122 | mask=False, |
| 6123 | fill_value=999999) |
| 6124 | |
| 6125 | >>> x.ptp(axis=0) |
| 6126 | masked_array(data=[2, 0, 5, 2], |
| 6127 | mask=False, |
| 6128 | fill_value=999999) |
| 6129 | |
| 6130 | >>> x.ptp() |
| 6131 | 10 |
| 6132 | |
| 6133 | This example shows that a negative value can be returned when |
| 6134 | the input is an array of signed integers. |
| 6135 | |
| 6136 | >>> y = np.ma.MaskedArray([[1, 127], |