log-det = log|abs(|W|)| * pixels

args: hidden_state:list of tuples, one for every layer, each tuple should be hidden_layer_i,c_layer_i input is the te

[summary] Gets an angle for each step [description] The maximal angle should be larger the more intens

[summary] Here we calculate the length of the steps taken by the motion blur [description] We want a h

get image path list support lmdb or image files

Generate an index list for reading N frames from a sequence of images Args: crt_i (int): current center index max_n (int): max num

[summary] Kernel matrix of motion blur of given intensity. [description] Once generated, it stays the same. Returns:

pytorch torch.nn.Linear wrapper Notes --------------------------------------------------------------------- Arguments: activation_

img_in: Numpy, HWC or HW

[summary] L^2 norm of a list [description] Used for internals Arguments: lst {list} -- vector

Read a sequence of images from a given folder path Args: path (list/str): list of image paths/image folder path Returns: imgs

same as matlab rgb2ycbcr only_y: only return Y channel Input: uint8, [0, 255] float, [0, 1]

set up logger

Compute softmax values for each sets of scores in x.