add new dementions at the front and the tail

r"""Create 1-D gauss kernel Args: size (int): the size of gauss kernel sigma (float): sigma of normal distribution Returns:

r""" Calculate ssim index for X and Y Args: X (torch.Tensor): images Y (torch.Tensor): images win (torch.Tensor): 1-D gaus

sum over the first and last dimention

Create Augmented Dataset

Args: image: np.ndarray, shape: C X H X W

Execute an replication callback `__data_parallel_replicate__` on each module created by original replication. The callback will be invoked w

Construct Volumes from Original High Dimensional (D) Data

r""" class for ms-ssim Args: win_size: (int, optional): the size of gauss kernel win_sigma: (float, optional): sigma o

Compute the mean and standard-deviation with sum and square-sum. This method also maintains the moving average on the master device.

r""" Blur input with 1-D kernel Args: input (torch.Tensor): a batch of tensors to be blured window (torch.Tensor): 1-D gauss kerne

r""" interface of ms-ssim Args: X (torch.Tensor): a batch of images, (N,C,H,W) Y (torch.Tensor): a batch of images, (N,C,H,W)

Register an slave device. Args: identifier: an identifier, usually is the device id. Returns: a `SlavePipe` obj

Main entry for the master device in each forward pass. The messages were first collected from each devices (including the master devi

r""" interface of ssim Args: X (torch.Tensor): a batch of images, (N,C,H,W) Y (torch.Tensor): a batch of images, (N,C,H,W)

Warning: this code is not compatible with bandwise flag

Args: img (np.mdarray): Image to be geometric transformed. Returns: np.ndarray: Randomly geometric transform