Metadata-Version: 2.1
Name: ndtransform
Version: 0.1.2
Summary: 
Author: sevolek
Author-email: ale.sev@outlook.com
Requires-Python: >=3.11,<4.0
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Requires-Dist: numpy (>=1.24.3,<2.0.0)
Description-Content-Type: text/markdown

This library contains 3 functions that are meant to be used for tensor transformations. 
Mostly they are built using standard python library with some functionalities of numpy.

The 3 functions in question:

-transpose2d(tensor): takes a tensor in the format "list[list[float]]". This tensor will
then be transposed. Returns a list.

-window1d(array, size, *shift, *stride): "array" takes an array in format list or np.ndarray;
"size", of which size you want to get your windows; "shift" (default = 1) optional argument, which 
determines step in between window calculations; "stride" (default = 1) optional argument, which 
represents the step size. Returns a list.

-convolution2d(matrix, kernel, *stride): "matrix" takes an np.ndarray that you would like to make 
calculations on; "kernel" takes an np.ndarray that will be used to do calculations on original
matrix; "stride" (default = 1) optional argument that represents the step size, when generating
result. Returns np.ndarray.
