pyiqa.archs.stlpips_arch
========================

.. py:module:: pyiqa.archs.stlpips_arch

.. autoapi-nested-parse::

   ST-LPIPS Model

   github repo link: https://github.com/abhijay9/ShiftTolerant-LPIPS

   Cite as:
   @inproceedings{ghildyal2022stlpips,
       title={Shift-tolerant Perceptual Similarity Metric},
       author={Abhijay Ghildyal and Feng Liu},
       booktitle={European Conference on Computer Vision},
       year={2022}
   }









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

.. py:data:: default_model_urls

.. py:function:: spatial_average(in_tens, keepdim=True)

.. py:function:: upsample(in_tens, out_HW=(64, 64))

.. py:function:: normalize_tensor(in_feat, eps=1e-10)

.. py:class:: STLPIPS(pretrained=True, net='alex', variant='shift_tolerant', lpips=True, spatial=False, pnet_tune=False, use_dropout=True, pretrained_model_path=None, eval_mode=True, blur_filter_size=3)

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


   ST-LPIPS model.
   :param lpips: Whether to use linear layers on top of base/trunk network.
   :type lpips: Boolean
   :param pretrained: Whether means linear layers are calibrated with human
                      perceptual judgments.
   :type pretrained: Boolean
   :param net: ['alex','vgg','squeeze'] are the base/trunk networks available.
   :type net: String
   :param pretrained_model_path: Petrained model path.
   :type pretrained_model_path: String
   :param The following parameters should only be changed if training the network:
   :param eval_mode: choose the mode; True is for test mode (default).
   :type eval_mode: Boolean
   :param pnet_tune: Whether to tune the base/trunk network.
   :type pnet_tune: Boolean
   :param use_dropout: Whether to use dropout when training linear layers.
   :type use_dropout: Boolean


   .. py:method:: forward(in0, in1, retPerLayer=False, normalize=True)

      Computation IQA using LPIPS.
      :param in1: An input tensor. Shape :math:`(N, C, H, W)`.
      :param in0: A reference tensor. Shape :math:`(N, C, H, W)`.
      :param retPerLayer: return result contains result of
                          each layer or not. Default: False.
      :type retPerLayer: Boolean
      :param normalize: Whether to normalize image data range
                        in [0,1] to [-1,1]. Default: True.
      :type normalize: Boolean

      :returns: Quality score.



.. py:class:: ScalingLayer

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


   .. py:method:: forward(inp)


.. py:class:: NetLinLayer(chn_in, chn_out=1, use_dropout=False)

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


   A single linear layer which does a 1x1 conv


   .. py:method:: forward(x)


.. py:class:: alexnet(requires_grad=False, variant='shift_tolerant', filter_size=3)

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


   .. py:method:: forward(X)


.. py:class:: vggnet(requires_grad=False, variant='shift_tolerant', filter_size=3)

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


   .. py:method:: forward(X)


.. py:class:: Downsample(pad_type='reflect', filt_size=3, stride=2, channels=None, pad_off=0, pad_size='', pad_more=False)

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


   .. py:method:: forward(inp)


.. py:function:: get_pad_layer(pad_type)

.. py:class:: VGG(features, num_classes=1000, init_weights=True)

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


   .. py:method:: forward(x)


.. py:function:: make_layers(cfg, batch_norm=False, filter_size=1, pad_more=False, fconv=False)

.. py:data:: cfg

.. py:function:: vgg16(pretrained=False, filter_size=1, pad_more=False, fconv=False, **kwargs)

   VGG 16-layer model (configuration "D")

   :param pretrained: If True, returns a model pre-trained on ImageNet
   :type pretrained: bool