Utils

get_func_name(fun)

Returns the name of a function passed as parameter being either a function itself or a string with the function name:

from cvnn.utils import get_func_name

get_func_name(print)    # output: 'print'
get_func_name('print')  # output: 'print'
Parameters:fun – function or function name
Returns:function name
create_folder(root_path, now=None)

Creates folders within root_path using a date format.

Parameters:
  • root_path – root path where to create the folder chain
  • now – date to be used. If None then it will use current time
Returns:

the created path in pathlib format (compatible across different OS)
transform_to_real(x_complex, mode="real_imag")
Transforms a complex input matrix into a real value matrix (double size)
Parameters:
  • x_complex – Complex-valued matrix of size mxn
  • mode –

    Mode on how to transform to real. One of the following.

    • real_imag (default): Separate x_complex into real and imaginary making the size of the return double x_complex
    • amplitude_phase: Separate x_complex into amplitude and phase making the size of the return double x_complex
    • amplitude_only: Apply the absolute value to x_complex. Shape remains the same.
Returns:

real-valued matrix of real valued cast of x_complex
randomize(x, y):

Randomizes the order of data samples and their corresponding labels

Parameters:
  • x – data
  • y – data labels
Returns:

Tuple of (shuffled_x, shuffled_y) maintaining coherence of elements labels
polar2cart(rho, angle):
\[z = \rho \cdot e^{j\phi}\]
Parameters:
  • rho – absolute value
  • angle – phase
Returns:

complex number using phase and angle
cart2polar(z):
Parameters:z – complex input
Returns:tuple with the absolute value of the input and the phase