Transpose

class deeptrack.features.Transpose(axes: tuple[int, ...] | None = None, **kwargs: Any)

Bases: Feature

Transpose the input array or tensor.

This feature rearranges the axes of an input array or tensor according to the specified order. The axes parameter determines the new order of the dimensions.

Parameters

axes: tuple[int, …], optional

A tuple specifying the permutation of the axes. If None (default), the axes are reversed.

**kwargs: Any

Additional keyword arguments passed to the parent Feature class.

Methods

get(inputs, axes, **kwargs) -> array or tensor

Transpose the axes of the input array(s) or tensor(s). The inputs and outputs can be NumPy arrays or PyTorch tensors.

Examples

>>> import deeptrack as dt

Create an input array:

>>> import numpy as np
>>>
>>> input_array = np.random.rand(2, 3, 4)
>>> input_array.shape
(2, 3, 4)

Apply a Transpose feature:

>>> transpose_feature = dt.Transpose(axes=(1, 2, 0))
>>> output_array = transpose_feature(input_array)
>>> output_array.shape
(3, 4, 2)

Without specifying axes:

>>> transpose_feature = dt.Transpose()
>>> output_array = transpose_feature(input_array)
>>> output_array.shape
(4, 3, 2)

Methods Summary

get(inputs[, axes])

Transpose the axes of the input array or tensor.

Methods Documentation

get(inputs: ndarray | Tensor, axes: tuple[int, ...] | None = None, **kwargs: Any) → ndarray | Tensor

Transpose the axes of the input array or tensor.

Parameters

inputs: array or tenor

The input array or tensor to process. The input can be a NumPy array or a PyTorch tensor.

axes: tuple[int, …], optional

A tuple specifying the permutation of the axes. If None (default), the axes are reversed.

**kwargs: Any

Additional keyword arguments (unused here).

Returns

array or tensor

The transposed image with rearranged axes. The output can be a NumPy array or a PyTorch tensor.