de-transformation 0.1.5

Advanced Python & Linux Shell Commands: Project

Python package project

Description

This project has a python package de-transformation code to make data transformations. The package consists of 3 functions:

• transpose2d
• window1d
• convolution2d

transpose2d

Description

Transposes a 2D matrix.

Parameters

• input_matrix (list[list[float]]): The input 2D matrix to be transposed.

Return Value

• list[list[float]]: The transposed 2D matrix, where rows become columns and columns become rows.

Example

from de_transformation.utils import transpose2d

# Example input matrix
input_matrix = [[1.0, 2.0, 3.0],
                [4.0, 5.0, 6.0]]

# Transpose the matrix
transposed_matrix = transpose2d(input_matrix)

# Output:
# transposed_matrix is now:
# [[1.0, 4.0],
#  [2.0, 5.0],
#  [3.0, 6.0]]

window1d

Description

Extracts window sub-arrays from a 1D list or numpy array, with the option to specify the size of the windows, the shift for the starting position of the window, and the stride between consecutive windows.

Parameters

• input_array (list or np.ndarray): The input 1D array or list from which windows are extracted.
• size (int): The size of the windows to extract.
• shift (int, optional): The number of elements to shift the starting position of the window (default is 1).
• stride (int, optional): The step size between consecutive windows (default is 1).

Return Value

• list of lists or np.ndarrays: A list containing windows of the specified size extracted from the input array.

Example

from de_transformation.utils import window1d

# Example input array
input_array = [1, 2, 3, 4, 5, 6, 7, 8, 9]

# Extract windows of size 3 with a shift of 2
windows = window1d(input_array, size=3, shift=2)

# Output:
# windows is now:
# [[1, 2, 3],
#  [3, 4, 5],
#  [5, 6, 7],
#  [7, 8, 9]]

convolution2d

Description

Performs 2D convolution on a numpy array using a specified convolution kernel and stride.

Parameters

• input_matrix (numpy.ndarray): The input 2D matrix to be convolved.
• kernel (numpy.ndarray): The convolution kernel (filter) to apply.
• stride (int, optional): The stride for the convolution operation (default is 1).

Return Value

• numpy.ndarray: The result of the 2D convolution operation, which is a numpy array.

Usage Example

from de_transformation.utils import convolution2d

# Example input matrix
input_matrix = np.array([[1, 2, 3],
                         [4, 5, 6],
                         [7, 8, 9]])

# Example kernel
kernel = np.array([[1, 0],
                   [0, -1]])

# Perform 2D convolution
result = convolution2d(input_matrix, kernel)

# Output:
# result is now:
# [[  1.  -2.]
#  [ -1.  -2.]]