A Dataset of the low and high light images with data values in each channel in the range 0-1 (normalized).

Similar to nn.Sequential, but it lets us introduce a second argument in the forward method so adaptors can be considered in the inference.

Change this block using Branch

Branch that lasts lonly the dilated convolutions

Change this block using Branch

Charbonnier loss (one variant of Robust L1Loss, a differentiable variant of L1Loss). Described in "Deep Laplacian Pyramid Networks for Fast a

A function that crops an image into 4 patches (top-left, top-right, bottom-left, bottom-right), adn returns a list of the patches.

Applies the enhanceLoss. This loss is the l1 loss of the image downsampled at the middle of the encoder-decoder plus the l1 of the features o

Calculates the amplitude of frequencies loss.

L1 (mean absolute error, MAE) loss. Args: loss_weight (float): Loss weight for L1 loss. Default: 1.0. reduction (str): Specifies

MSE (L2) loss. Args: loss_weight (float): Loss weight for MSE loss. Default: 1.0. reduction (str): Specifies the reduction to app

A function that random crops a pair of images with a fixed size (and the same crop).

L1 Charbonnierloss.

L_deblur.

L_enhance.

L_enhance.