Forward function.

Perform the computation Parameters: outputs: raw outputs of the model target_sizes: tensor of dimension [batch_size x

query: (n_context, queryL, d) context: (n_context, sourceL, d)

Generalized IoU from https://giou.stanford.edu/ The boxes should be in [x0, y0, x1, y1] format Returns a [N, M] pairwise matrix, where

Generalized IoU from https://giou.stanford.edu/ Input: - boxes1, boxes2: N,4 Output: - giou: N, 4

Generate attention mask between each pair of special tokens Args: input_ids (torch.Tensor): input ids. Shape: [bs, num_token] spec

get_gaussian_map_from_points B,C,H,W -> B,N,2 float(0, 1) float(0, 1) softargmax function Args: hm (float): Input images(

return the dicf contained in args. e.g: >>> with open(path, 'w') as f: json.dump(get_raw_dict(args), f, indent=2)

L1-normalize columns of X

Compute the bounding boxes around the provided masks The masks should be in format [N, H, W] where N is the number of masks, (H, W) are the spati

Merge list into cfg_dict Merge the dict parsed by MultipleKVAction into this cfg. Examples: >>> options = {'model.backbo

import cv2 image = cv2.imread(IMAGE_PATH) model = Model(model_config_path=CONFIG_PATH, model_checkpoint_path=WEIGHTS_PATH)

import cv2 image = cv2.imread(IMAGE_PATH) model = Model(model_config_path=CONFIG_PATH, model_checkpoint_path=WEIGHTS_PATH)

Take a list of strings ``ls`` and colors ``lc`` and place them next to each other, with text ls[i] being shown in color lc[i]. This exam

Simple version of ``kwimage.Boxes.random`` Returns: Tensor: shape (n, 4) in x1, y1, x2, y2 format. References: https://gitla

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

Register a module build function. Args: module (:obj:`nn.Module`): Module to be registered.

Initialize the detectron2 logger and set its verbosity level to "INFO". Args: output (str): a file name or a directory to save log.

Display the specified annotations. :param anns (array of object): annotations to display :return: None

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

Dump data to json/yaml/pickle strings or files. This method provides a unified api for dumping data as strings or to files, and also supports

Warning: does not synchronize the deque!

Moves the target dicts to the given device.

remove the padding and convert to img list Returns: [type]: [description]

Convert the model into training mode while keep layers freezed.

Input: pred, gt: Tensor()

return if the results is the best.

img: tensor(3, H, W) tgt: make sure they are all on cpu. must have items: 'image_id', 'boxes', 'size'