Degradation pipeline, modified from Real-ESRGAN: https://github.com/xinntao/Real-ESRGAN

Make dirs for experiments.

Support OrderedDict for yaml. Returns: tuple: yaml Loader and Dumper.

Generate paired paths from folders. Args: folders (list[str]): A list of folder path. The order of list should be [input_fold

Generate paired paths from lmdb files. Contents of lmdb. Taking the `lq.lmdb` for example, the file structure is: :: lq.lmdb

Pixel unshuffle. Args: x (Tensor): Input feature with shape (b, c, hh, hw). scale (int): Downsample ratio. Returns:

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

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

Save networks and training state.

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

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

It will first crop input images to tiles, and then process each tile. Finally, all the processed tiles are merged into one images. Mo

Apply wavelet decomposition to the input tensor. This function only returns the low frequency & the high frequency.

Args: x: (b, h, w, c) window_size (int): window size Returns: windows: (num_windows*b, window_size, window_size, c)

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

It is the training pair pool for increasing the diversity in a batch. Batch processing limits the diversity of synthetic degradations in a ba

It is the training pair pool for increasing the diversity in a batch. Batch processing limits the diversity of synthetic degradations in a ba

It is the training pair pool for increasing the diversity in a batch. Batch processing limits the diversity of synthetic degradations in a ba

It is the training pair pool for increasing the diversity in a batch, taken from Real-ESRGAN: https://github.com/xinntao/Real-ESRGAN

Take the prediction of fake and real images from the combined batch. The prediction contains the intermediate outputs of multiscale G

Generates a gaussian mask of weights for tile contributions

Get the initial lr, which is set by the scheduler.

Initialize slurm distributed training environment. If argument ``port`` is not specified, then the master port will be system environment var