Convert a recursive dictionary into a flat, dot-map dictionary. :param d: e.g. {'a': {'b': 2, 'c': 3}} :param parent_key: Used for recur

Update network weights; it will be called in every training iteration.

Update network weights; it will be called in every training iteration.

display the current losses on visdom display: dictionary of error labels and values Parameters: epoch (int) -- current

Draw a grid showing how ground truth objects are classified. gt_class_ids: [N] int. Ground truth class IDs pred_class_id: [N] int. Predicted c

Draw the precision-recall curve. AP: Average precision at IoU >= 0.5 precisions: list of precision values recalls: list of recall values

print current losses on console; also save the losses to the disk Parameters: epoch (int) -- current epoch iters (int

Print the mean, min, max, median, std, and size of a numpy array Parameters: val (bool) -- if print the values of the numpy array

Remove anything printed in the last printover call. Then print `string`

Return an image from the pool. Parameters: images: the latest generated images from the generator Returns images from th

Get all (key, item) recursively in a potentially recursive dictionary. Usage: ``` x = { 'foo' : { 'bar' : 5

Args: input_dict (dict): all the values will be reduced average (bool): whether to do average or sum Reduce the values in the

Reset the self.saved status

Resize masks size :param masks: numpy of shape (n, 1, h, w) :param image_size: H, W :return: numpy array of shape (n, H, W)

Draw a rounded rectangle

Save a numpy image to the disk Parameters: image_numpy (numpy array) -- input numpy array image_path (str) -- the path o

Save images to the disk. Parameters: webpage (the HTML class) -- the HTML webpage class that stores these imaegs (see html.py for more de

Save all the networks to the disk. Parameters: epoch (int) -- current epoch; used in the file name '%s_net_%s.pth' % (epoch, name

Save all the networks to the disk. Parameters: epoch (int) -- current epoch; used in the file name '%s_net_%s.pth' % (epoch, name

Unpack input data from the dataloader and perform necessary pre-processing steps. Parameters: input: a dictionary that contains t

Unpack input data from the dataloader and perform necessary pre-processing steps. Parameters: input: a dictionary that contains t

Set requies_grad=Fasle for all the networks to avoid unnecessary computations Parameters: nets (network list) -- a list of netwo

Args: worker: a function with argument rank, world_size, args_in example see test_ddp_spawn Returns:

Warning: does not synchronize the deque!

Get `num` elements from the sequence that are as spread out as possible. https://stackoverflow.com/questions/9873626/choose-m-evenly-spaced-

r"""Convert flow tensor to color image. Args: tensor (tensor) of If tensor then (NxCxHxW) or (NxTxCxHxW) or (CxHxW). imty

Args: image_tensor: (list of) shape [..., C, H, W] Returns:

Converts a Tensor array into a numpy image array. Parameters: input_image (tensor) -- the input image tensor array imtype (type)

r"""Simple function wrapper for the master function

Write optical flow to file. If v is None, uv is assumed to contain both u and v channels, stacked in depth. Original code by Deqing Sun,