pyiqa.archs.entropy_arch
========================

.. py:module:: pyiqa.archs.entropy_arch

.. autoapi-nested-parse::

   Entropy Metric for grayscale images.
   Entropy is a statistical measure of randomness that can be used to characterize the texture of the input image.

   Created by: Chaofeng Chen (https://github.com/chaofengc)

   Refer to:
       Matlab: https://www.mathworks.com/help/images/ref/entropy.html







Module Contents
---------------

.. py:function:: entropy(x, data_range=255.0, eps=1e-08, color_space='yiq')

   Compute grayscale entropy from an image tensor.

   :param x: Input tensor with shape ``(N, C, H, W)``.
   :type x: torch.Tensor
   :param data_range: Maximum intensity value used for histogram bins.
   :type data_range: float
   :param eps: Numerical stability constant in ``log2``.
   :type eps: float
   :param color_space: Color space used when converting RGB to luminance.
   :type color_space: str

   :returns: Entropy values with shape ``(N,)``.
   :rtype: torch.Tensor


.. py:class:: Entropy(**kwargs)

   Bases: :py:obj:`torch.nn.Module`


   Entropy-based no-reference image quality metric wrapper.

   :param \*\*kwargs: Keyword arguments forwarded to :func:`entropy`.


   .. py:method:: forward(x)

      Compute entropy score for input batch.

      :param x: Image tensor with shape ``(N, C, H, W)``.
      :type x: torch.Tensor

      :returns: Entropy score tensor with shape ``(N,)``.
      :rtype: torch.Tensor