Return a batch of data

Return the total number of images in the dataset.

Return the number of data in the dataset

Modified focal loss. Exactly the same as CornerNet. Runs faster and costs a little bit more memory Arguments: pred (batch x c x

focal loss from CornerNet

Run all_gather on arbitrary picklable data (not necessarily tensors) Args: data: any picklable object Returns: list[data]

Split an interable into batches of size `n`. If `n` does not evenly divide `iterable`, the last slice will be smaller. https://stackover

assuming each tensor in the list has shape [brick_num, ...], and len(l) = batch_size we build a tensor of [ba

assuming each tesnor in the list has shape [brick_num, ...], and len(l) = batch_size we build a tensor of [batch_size, self.n

Clip the magnitude of value to be within some value. :param value: :param magnitude: :return:

Args: tensor: expect range [0, 1] pad_value: padding: process_grid: e.g. tensor2im, tensor2flow value_ch

r"""Evaluates module(input) in parallel across the GPUs given in device_ids. This is the functional version of the DataParallel module. Args

Calculate and print the mean of average absolute(gradients) Parameters: net (torch network) -- Torch network name (str) -- the na

``` x = {'foo': [3, 4, 5], 'bar': [1, 2, 3]} ppp.dict_of_list__to__list_of_dicts(x, 3) # Output: # [ # {'foo': 3, 'bar':

Convert each value in the dictionary into a JSON'able primitive. :param d: :return:

Args: layout: a *2D* list, each element is a dictionary with keys 'info' and 'image' epoch: iter:

Display ground truth and prediction instances on the same image.

Display the given image and the top few class masks.

Scans all the weights in the model and returns a list of tuples that contain stats about each weight.

r""" gather to all ranks

r""" Reduce to all ranks

r""" Reduce to rank 0

Convert something like ``` { 'one.two.three.four': 4, 'one.six.seven.eight': None, 'five.nine.ten': 10, '

Draw 3-pixel width bounding boxes on the given image array. color: list of 3 int values for RGB.

Draw bounding boxes and segmentation masks with different customizations. boxes: [N, (y1, x1, y2, x2, class_id)] in image coordinates. re

anchors: [n, (y1, x1, y2, x2)] list of anchors in image coordinates. proposals: [n, 4] the same anchors but refined to fit objects better.

Filter out elements that are Falsy (where bool(x) is False) from potentially recursive lists. :param x_or_iterable: An element or a list