Source code for draugr.torch_utilities.operations.sizes.conv2d

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

__author__ = "Christian Heider Nielsen"
__doc__ = r"""

           Created on 23/07/2020
           """

import math
from typing import Tuple, Union

from warg import replicate

__all__ = ["conv2d_padding_size", "conv2d_hw_shape"]


[docs]def conv2d_hw_shape( h_w: Union[int, Tuple[int, int]], kernel_size: Union[int, Tuple[int, int]] = 1, stride: Union[int, Tuple[int, int]] = 1, pad: Union[int, Tuple[int, int], Tuple[int, int, int, int]] = 0, dilation: Union[int, Tuple[int, int]] = 1, ) -> Tuple[int, int]: """ :param h_w: :type h_w: :param kernel_size: :type kernel_size: :param stride: :type stride: :param pad: :type pad: :param dilation: :type dilation: :return: :rtype:""" (h_w, kernel_size, stride, pad, dilation) = ( replicate(h_w), replicate(kernel_size), replicate(stride), replicate(pad), replicate(dilation), ) pad = (replicate(pad[0]), replicate(pad[1])) h = math.floor( (h_w[0] + sum(pad[0]) - dilation[0] * (kernel_size[0] - 1) - 1) / stride[0] + 1 ) w = math.floor( (h_w[1] + sum(pad[1]) - dilation[1] * (kernel_size[1] - 1) - 1) / stride[1] + 1 ) return h, w
[docs]def conv2d_padding_size( h_w_in: Union[int, Tuple[int, int]], h_w_out: Union[int, Tuple[int, int]], kernel_size: int = 1, stride: int = 1, dilation: int = 1, ) -> Tuple[Tuple[int, int], Tuple[int, int]]: """ :param h_w_in: :type h_w_in: :param h_w_out: :type h_w_out: :param kernel_size: :type kernel_size: :param stride: :type stride: :param dilation: :type dilation: :return: :rtype:""" (h_w_in, h_w_out, kernel_size, stride, dilation) = ( replicate(h_w_in), replicate(h_w_out), replicate(kernel_size), replicate(stride), replicate(dilation), ) p_h = ( (h_w_out[0] - 1) * stride[0] - h_w_in[0] + dilation[0] * (kernel_size[0] - 1) + 1 ) p_w = ( (h_w_out[1] - 1) * stride[1] - h_w_in[1] + dilation[1] * (kernel_size[1] - 1) + 1 ) return ( (math.floor(p_h / 2), math.ceil(p_h / 2)), (math.floor(p_w / 2), math.ceil(p_w / 2)), )
if __name__ == "__main__": print(conv2d_hw_shape(105, (0, 1, 2, 3)))