Build an conv block for IndexBlock. Args: in_channels (int): The input channels of the block. out_channels (int): The output chan

Build multiple optimizers from configs. If `cfgs` contains several dicts for optimizers, then a dict for each constructed optimizers will be

Returns list of utf-8 byte and a corresponding list of unicode strings. The reversible bpe codes work on unicode strings. This means y

Calculate multiple types of losses. Args: loss_type (str): Type of the loss. fake_res (torch.Tensor): Direct results

Calculate multiple types of losses. Args: loss_type (str): Type of the loss. fake_res (torch.Tensor): Direct results

Charbonnier loss. Args: pred (Tensor): Prediction Tensor with shape (n, c, h, w). target ([type]): Target Tensor with shape (n, c

Return intensity limits, i.e. (min, max) tuple, of the image's dtype. This function is adopted from skimage: https://github.com/scikit-image/

Estimate AGGD (Asymmetric Generalized Gaussian Distribution) parameters. Args: block (ndarray): 2D Image block. Returns: tup

Evaluation function. Args: output (Tensor): Model output. target (Tensor): GT Tensor. Returns: d

Extract patches around the given bbox. Args: bbox (np.ndarray | torch.Tensor): Bboxes to be modified. Bbox can be in batch or

Forward function. Args: img_a (Tensor): Input image from domain A. img_b (Tensor): Input image from domain B.

Forward function. Args: img_a (Tensor): Input image from domain A. img_b (Tensor): Input image from domain B.

Forward Function. Args: pred (Tensor): of shape (N, C, H, W). Predicted tensor. target (Tensor): of shape (N, C, H, W

Forward of networks. Args: lq (Tensor): LQ image. lq_up (Tensor): Upsampled LQ image. ref (Tensor): Refer

Abstract method for testing forward. All subclass should overwrite it.

Forward function for testing. Args: masked_img (torch.Tensor): Tensor with shape of (n, 3, h, w). mask (torch.Tensor)

Testing forward function. This is a basic function, interpolate a frame between the given two frames. Args: inpu

Testing forward function. Args: lq (Tensor): LQ image. gt (Tensor): GT image. meta (list[dict]): Meta dat

Testing forward function. Args: lq (Tensor): LQ Tensor with shape (n, c, h, w). gt (Tensor): GT Tensor with shape (n,

Forward function in discriminator training step. In this function, we modify the default implementation with only one discriminator.

Forward function in discriminator training step. In this function, we compute the prediction for each data batch (real or fake). Mean

Gaussian gradient. From https://www.mathworks.com/matlabcentral/mlc-downloads/downloads/ submissions/8060/versions/2/previews/gaussgradient/g

Gaussian function. Args: x (array_like): The independent variable. sigma (float): Standard deviation of the gaussian function.

Forward function in generator training step. In this function, we mainly compute the loss items for generator with the given (fake_re

Get the Gaussian filter coefficients in one dimension (x or y direction). Args: kernel_size (int): Kernel filter size in

Get an inner module from model. Since we will wrapper DDP for some model, we have to judge whether the module can be indexed directly

Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).

Get target label. Args: input (Tensor): Input tensor. target_is_real (bool): Whether the target is real or fake.

Init weights for models. Args: pretrained (str, optional): Path for pretrained weights. If given None, pretrained

Load unpaired image paths of one domain. Args: dataroot (str): Path to the folder root for unpaired images of one

Make coordinates at grid centers. Args: shape (tuple): shape of image. ranges (tuple): range of coordinate value. Default: None.

MSE loss. Args: pred (Tensor): Prediction Tensor with shape (n, c, h, w). target ([type]): Target Tensor with shape (n, c, h, w).

Test model with multiple gpus. This method tests model with multiple gpus and collects the results under two different modes: gpu and cpu mod

nn.Module: normalization layer after the second convolution layer

Query current image batch using a history of generated images. Args: images (Tensor): Current image batch without history informa

Query RGB value of GT. Adapted from 'https://github.com/yinboc/liif.git' 'liif/models/liif.py' Copyright (c) 2020, Yinbo Chen

Randomly generate bivariate isotropic or anisotropic Gaussian kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and

Randomly generate bivariate generalized Gaussian kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and `rotation_

Randomly generate bivariate plateau kernels. In the isotropic mode, only `sigma_x_range` is used. `sigma_y_range` and `rotation_range` is ign

Randomly choose an unknown start (top-left) point for a given crop_size. Args: unknown (np.ndarray): The binary unknown mask. cro

Generate random irregular masks. This is a modified version of free-form mask implemented in 'brush_stroke_mask'. We prefer to use `uint

Read image from file. Args: filepath (str): File path. Returns: image (np.array): Image.

Resize the given image to a given size. Args: img (ndarray | torch.Tensor): The input image. size (int | tuple[int]): Target size

224 288 384

Set random seed. Args: seed (int): Seed to be used. deterministic (bool): Whether to set the deterministic option for

Perform necessary pre-processing steps. Args: img_a (Tensor): Input image from domain A. img_b (Tensor): Input image

Perform necessary pre-processing steps. Args: img_a (Tensor): Input image from domain A. img_b (Tensor): Input image

Test model with a single gpu. This method tests model with a single gpu and displays test progress bar. Args: model (nn.Module): Mod

Apply spatial ensemble. Args: imgs (torch.Tensor): The images to be processed by the model. Its size should be ei

Concat datasets with different ann_file but the same type. Args: cfg (dict): The config of dataset. default_args (dict, optional)

Distributed training function. Args: model (nn.Module): The model to be trained. dataset (:obj:`Dataset`): Train dataset.

Freeze stages param and norm stats.

Generate One Heatmap. Args: landmark (Tuple[float]): Location of a landmark. results: heatmap (np.ndarray):

Helper method to compute inverse matrix for affine transformation. As it is explained in PIL.Image.rotate, we need compute INVERSE of

Initialize conv weights as described in PGGAN.