Get environment information. Currently, only log the software version.

Get environment information. Currently, only log the software version.

Calculate quality scores for an image using multiple metrics. Args: source_image_path (str): Path to the image.

Get the position from a period list. It will return the index of the right-closest number in the period list. For example, the cumulative_per

Gets the intermediate features from the ResNet50 backbone. Args: x (torch.Tensor): The input tensor. Returns:

Gets the intermediate features from the Vision Transformer backbone. Args: x (torch.Tensor): The input tensor.

Load network. Args: load_path (str): The path of networks to be loaded. net (nn.Module): Network. strict

Load S2Former model weights Args: model_path (str): Path to model weights device (torch.device): Device to load model on

Make dirs for experiments.

Make dirs for experiments.

Make dirs for experiments.

mkdirs. If path exists, rename it with timestamp and create a new one. Args: path (str): Folder path.

r"""Calculate nancov for batched tensor, rows that contains nan value will be removed. Args: x (tensor): (B, row_num, feat_dim)

r"""nanmean same as matlab function: calculate mean values by removing all nan.

Parse option file. Args: opt_path (str): Option file path. is_train (str): Indicate whether in training or not. Default: True.

Generate paths from lmdb. Args: folder (str): Folder path. Returns: list[str]: Returned path list.

Pre-process, such as pre-pad and mod pad, so that the images can be divisible

Default BGR in [0, 255] in original codes

Preprocesses the input tensor by normalizing it. Args: x (torch.Tensor): The input tensor. Returns:

Process image using the tool engine and restore to original size Args: image_path (str): Path to the input image models

Calculate factor corresponding to quality Args: quality(float): Quality for jpeg compression. Returns: float: Compression f

Randomly crops the input tensor(s) to the specified size and number of crops. Args: - input_list (list or tensor): List of input ten

reduce loss dict. In distributed training, it averages the losses among different GPUs . Args: loss_dict (OrderedDict):

Reflects an array around two points making a triangular waveform that ramps up and down, allowing for pad lengths greater than the input len

Register the given object under the the name `obj.__name__`. Can be used as either a decorator or not. See docstring of this

Computes the quality score for a distorted and reference image pair. Args: - dis (torch.Tensor): The distorted image.

Regress the score for the input image. Args: dis (torch.Tensor): The distorted image. ref (torch.Tensor): Th

Args: x (torch.Tensor): A torch.Tensor of dimension (B x C, 1, H, W). dim (int): scale (float): size (int): R

r"""Convert a batch of RGB images to a batch of LMN images Args: x: Batch of images with shape (N, 3, H, W). RGB colour space. Retur

Rotates half the hidden dims of the input.

Perform S2Former prediction on a single image Args: model (torch.nn.Module): Loaded S2Former model input_img (str or PIL

Save image

Save networks. Args: net (nn.Module | list[nn.Module]): Network(s) to be saved. net_label (str): Network label.

Get all image files from a directory and return a sorted list of fullpath.

r"""Utility function that returns a normalized 3x3 Scharr kernel in X direction Returns: kernel: Tensor with shape (1, 3, 3)

r"""SDSP algorithm for salient region detection from a given image. Supports only colour images with RGB channel order. Args: x: Tenso

Set up schedulers.

Get human readable file size. Args: size (int): File size. suffix (str): Suffix. Default: 'B'. Return: str: Formatte

Get human readable file size. Args: size (int): File size. suffix (str): Suffix. Default: 'B'. Return: str: Formated

Get human readable file size. Args: size (int): File size. suffix (str): Suffix. Default: 'B'. Return: str: Formatte

Use SnowMaster model to perform snow removal on a single image Args: model (torch.nn.Module): Loaded model input_img (st