de2-1 0.3.1

Turing college de2_1 sprint submission. FINAl working version

This project contains the documentation for three functions: transpose2d, window1d, and convolution2d. These functions perform matrix transposition, sliding window generation, and 2D convolution, respectively.

de2_1 directory contains the project files, whereas tests contains the tests files. This library was built using poetry and published to pypi.

The project can be installed from here https://pypi.org/project/de2-1/

transpose2d

The transpose2d function transposes a 2D matrix by flipping its rows and columns. It takes a 2D list of floats as input and returns a 2D list of floats as output.

Arguments

• input_matrix (List[List[float]]): The input matrix to be transposed. It is expected to be a 2D list of floats.

Returns

• List[List[float]]: The transposed matrix as a 2D list of floats.

Examples

• python input_matrix = [[1, 2, 3], [4, 5, 6]]
• result = transpose2d(input_matrix) print(result) # [[1, 4], [2, 5], [3, 6]]

Notes

• The input_matrix should be a 2D list of floats.
• The function uses the zip(*input_matrix) expression to perform the transpose operation.
• The rows of the input_matrix become the columns of the transposed matrix, and vice versa.
• The resulting transposed matrix has dimensions (number of columns, number of rows) compared to the input matrix.
• The function returns the transposed matrix as a 2D list of floats.

window1d

The window1d function generates overlapping windows of a specified size from a 1D input array. It takes a 1D input array and optional parameters for window size, shift, and stride, and returns a list of windows, where each window is either a list or a 1D NumPy array.

Arguments

• input_array (List | np.ndarray): The 1D input array.
• size (int): The size of the window.
• shift (int, optional): The shift value between windows. Default is 1.
• stride (int, optional): The stride value for moving the window. Default is 1.

Returns

• List[Union[List, np.ndarray]]: A list of windows, where each window is either a list or a 1D NumPy array.

Raises

• TypeError: If the input_array is not a list or a 1D NumPy array.
• ValueError: If the size, shift, or stride is not a positive integer.

Examples

• python input_array = [1, 2, 3, 4, 5, 6, 7, 8, 9] size = 3 shift = 2 stride = 1
• result = window1d(input_array, size, shift, stride)
• print(result) # [[1, 2, 3], [3, 4, 5], [5, 6, 7], [7, 8, 9]]

Notes

• The input_array should be a 1D array, either a list or a 1D NumPy array.
• The size represents the length of each window.
• The shift determines the step size for moving the window starting position.
• The stride determines the step size for moving the window elements.
• If the window cannot fit entirely within the input_array, it is skipped.
• The function returns a list of windows, where each window is either a list or a 1D NumPy array.

convolution2d

The convolution2d function performs 2D convolution on an input matrix using a given kernel. at

Arguments

• input_matrix (List[List[float]]): The input matrix to be convolved. and is expected to be a 2D list of floats.
• kernel (List[List[float]]): The 2D kernel to be used for the convolution., and is expected to be a 2D list of floats.

Returns

• List[List[float]]: The convolved output matrix as a 2D list of floats.－

Examples

• python input_matrix = [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
• kernel = [[1, 0], [0, 1]]
• result = convolution2d(input_matrix, kernel)
• print(result) # [[5, 9], [16, 27]]

Notes

• The input_matrix should be a 2D list of floats. and the kernel should be a 2D list of floats. the convolved output matrix has the same number of rows as the input_matrix and one less column than the kernel.
• The function returns the convolved output matrix as a 2D list of floats.

Algorithm

1. Initialize an empty output matrix with the same number of rows as the input_matrix and one less column than the kernel.ommes. but not necessarily the same shape. The output matrix will have the same number of rows as the input_matrix and one less column than the kernel.
2. Iterate through each element in the input_matrix.
3. For each element, iterate through the kernel and apply the kernel to the corresponding element in the input_matrix.
4. Accumulate the results of the kernel application and store them in the output matrix.
5. Return the output matrix.