Caution

You're reading an old version of this documentation. If you want up-to-date information, please have a look at 0.9.1.

librosa.util.fix_length¶

librosa.util.fix_length(data, size, axis=- 1, **kwargs)[source]¶

Fix the length an array data to exactly size along a target axis.

If data.shape[axis] < n, pad according to the provided kwargs. By default, data is padded with trailing zeros.

Parameters

datanp.ndarray: array to be length-adjusted
sizeint >= 0 [scalar]: desired length of the array
axisint, <= data.ndim: axis along which to fix length
kwargsadditional keyword arguments: Parameters to np.pad

Returns

data_fixednp.ndarray [shape=data.shape]: data either trimmed or padded to length size along the specified axis.

See also

numpy.pad

Examples

>>> y = np.arange(7)
>>> # Default: pad with zeros
>>> librosa.util.fix_length(y, 10)
array([0, 1, 2, 3, 4, 5, 6, 0, 0, 0])
>>> # Trim to a desired length
>>> librosa.util.fix_length(y, 5)
array([0, 1, 2, 3, 4])
>>> # Use edge-padding instead of zeros
>>> librosa.util.fix_length(y, 10, mode='edge')
array([0, 1, 2, 3, 4, 5, 6, 6, 6, 6])