# `signal._arraytools`¶

Functions for acting on a axis of an array.

## Module Contents¶

### Functions¶

 `axis_slice`(a,start=None,stop=None,step=None,axis=None) Take a slice along axis ‘axis’ from ‘a’. `axis_reverse`(a,axis=None) Reverse the 1-d slices of a along axis axis. `odd_ext`(x,n,axis=None) Odd extension at the boundaries of an array `even_ext`(x,n,axis=None) Even extension at the boundaries of an array `const_ext`(x,n,axis=None) Constant extension at the boundaries of an array `zero_ext`(x,n,axis=None) Zero padding at the boundaries of an array
`axis_slice`(a, start=None, stop=None, step=None, axis=None)

Take a slice along axis ‘axis’ from ‘a’.

a : numpy.ndarray
The array to be sliced.
start, stop, step : int or None
The slice parameters.
axis : int, optional
The axis of a to be sliced.
```>>> a = array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
>>> axis_slice(a, start=0, stop=1, axis=1)
array([[1],
[4],
[7]])
>>> axis_slice(a, start=1, axis=0)
array([[4, 5, 6],
[7, 8, 9]])
```

The keyword arguments start, stop and step are used by calling slice(start, stop, step). This implies axis_slice() does not handle its arguments the exacty the same as indexing. To select a single index k, for example, use

axis_slice(a, start=k, stop=k+1)

In this case, the length of the axis ‘axis’ in the result will be 1; the trivial dimension is not removed. (Use numpy.squeeze() to remove trivial axes.)

`axis_reverse`(a, axis=None)

Reverse the 1-d slices of a along axis axis.

Returns axis_slice(a, step=-1, axis=axis).

`odd_ext`(x, n, axis=None)

Odd extension at the boundaries of an array

Generate a new ndarray by making an odd extension of x along an axis.

x : ndarray
The array to be extended.
n : int
The number of elements by which to extend x at each end of the axis.
axis : int, optional
The axis along which to extend x. Default is -1.
```>>> from scipy.signal._arraytools import odd_ext
>>> a = np.array([[1, 2, 3, 4, 5], [0, 1, 4, 9, 16]])
>>> odd_ext(a, 2)
array([[-1,  0,  1,  2,  3,  4,  5,  6,  7],
[-4, -1,  0,  1,  4,  9, 16, 23, 28]])
```

Odd extension is a “180 degree rotation” at the endpoints of the original array:

```>>> t = np.linspace(0, 1.5, 100)
>>> a = 0.9 * np.sin(2 * np.pi * t**2)
>>> b = odd_ext(a, 40)
>>> import matplotlib.pyplot as plt
>>> plt.plot(arange(-40, 140), b, 'b', lw=1, label='odd extension')
>>> plt.plot(arange(100), a, 'r', lw=2, label='original')
>>> plt.legend(loc='best')
>>> plt.show()
```
`even_ext`(x, n, axis=None)

Even extension at the boundaries of an array

Generate a new ndarray by making an even extension of x along an axis.

x : ndarray
The array to be extended.
n : int
The number of elements by which to extend x at each end of the axis.
axis : int, optional
The axis along which to extend x. Default is -1.
```>>> from scipy.signal._arraytools import even_ext
>>> a = np.array([[1, 2, 3, 4, 5], [0, 1, 4, 9, 16]])
>>> even_ext(a, 2)
array([[ 3,  2,  1,  2,  3,  4,  5,  4,  3],
[ 4,  1,  0,  1,  4,  9, 16,  9,  4]])
```

Even extension is a “mirror image” at the boundaries of the original array:

```>>> t = np.linspace(0, 1.5, 100)
>>> a = 0.9 * np.sin(2 * np.pi * t**2)
>>> b = even_ext(a, 40)
>>> import matplotlib.pyplot as plt
>>> plt.plot(arange(-40, 140), b, 'b', lw=1, label='even extension')
>>> plt.plot(arange(100), a, 'r', lw=2, label='original')
>>> plt.legend(loc='best')
>>> plt.show()
```
`const_ext`(x, n, axis=None)

Constant extension at the boundaries of an array

Generate a new ndarray that is a constant extension of x along an axis.

The extension repeats the values at the first and last element of the axis.

x : ndarray
The array to be extended.
n : int
The number of elements by which to extend x at each end of the axis.
axis : int, optional
The axis along which to extend x. Default is -1.
```>>> from scipy.signal._arraytools import const_ext
>>> a = np.array([[1, 2, 3, 4, 5], [0, 1, 4, 9, 16]])
>>> const_ext(a, 2)
array([[ 1,  1,  1,  2,  3,  4,  5,  5,  5],
[ 0,  0,  0,  1,  4,  9, 16, 16, 16]])
```

Constant extension continues with the same values as the endpoints of the array:

```>>> t = np.linspace(0, 1.5, 100)
>>> a = 0.9 * np.sin(2 * np.pi * t**2)
>>> b = const_ext(a, 40)
>>> import matplotlib.pyplot as plt
>>> plt.plot(arange(-40, 140), b, 'b', lw=1, label='constant extension')
>>> plt.plot(arange(100), a, 'r', lw=2, label='original')
>>> plt.legend(loc='best')
>>> plt.show()
```
`zero_ext`(x, n, axis=None)

Zero padding at the boundaries of an array

Generate a new ndarray that is a zero padded extension of x along an axis.

x : ndarray
The array to be extended.
n : int
The number of elements by which to extend x at each end of the axis.
axis : int, optional
The axis along which to extend x. Default is -1.
```>>> from scipy.signal._arraytools import zero_ext
>>> a = np.array([[1, 2, 3, 4, 5], [0, 1, 4, 9, 16]])
>>> zero_ext(a, 2)
array([[ 0,  0,  1,  2,  3,  4,  5,  0,  0],
[ 0,  0,  0,  1,  4,  9, 16,  0,  0]])
```