Efficient version of torch.cat that avoids a copy if there is only a single element in a list

Concatenates a list of BoxList (having the same image size) into a single BoxList Arguments: bboxes (list[BoxList])

Returns a resized copy of this bounding box :param size: The requested size in pixels, as a 2-tuple: (width, height).

From a set of original boxes and encoded relative box offsets, get the decoded boxes. Arguments: rel_codes (Tens

Helper function to synchronize between multiple processes when using distributed training

Compute the intersection over union of two set of boxes. The box order must be (xmin, ymin, xmax, ymax). Arguments: box1: (BoxList) bou

Encode a set of proposals with respect to some reference boxes Arguments: reference_boxes (Tensor): reference bo

Performs non-maximum suppression on a boxlist, with scores specified in a boxlist field via score_field. Arguments: boxlist(BoxL

tensors can be an ImageList, a torch.Tensor or an iterable of Tensors. It can't be a numpy array. When tensors is an iterable of Tensors,

Only keep boxes with both sides >= min_size Arguments: boxlist (Boxlist) min_size (int)

very similar to the smooth_l1_loss from pytorch, but with the extra beta parameter

Efficient version of torch.cat that avoids a copy if there is only a single element in a list

Given a vector of widths (ws) and heights (hs) around a center (x_ctr, y_ctr), output a set of anchors (windows).

Return width, height, x center, and y center for an anchor (window).

Simple function that adds fixed colors depending on the class

Generates a matrix of anchor boxes in (x1, y1, x2, y2) format. Anchors are centered on stride / 2, have (approximate) sqrt areas of the specified

Transpose bounding box (flip or rotate in 90 degree steps) :param method: One of :py:attr:`PIL.Image.FLIP_LEFT_RIGHT`, :py:

Arguments: in_channels_list (list[int]): number of channels for each feature map that will be fed out

Arguments: in_channels (int): number of channels of the input feature num_anchors (int): number of anchors to be pred

Arguments: in_channels (int): number of channels of the input feature num_anchors (int): number of anchors to be pred

Generate anchor (reference) windows by enumerating aspect ratios X scales wrt a reference (0, 0, base_size - 1, base_size - 1) window.

Enumerate a set of anchors for each aspect ratio wrt an anchor.

Enumerate a set of anchors for each scale wrt an anchor.

Arguments: proposals: list[BoxList] targets: list[BoxList]

Strategy: suppose that the models that we will create will have prefixes appended to each of its keys, for example due to an extra level of n

Arguments: dataset_list (list[str]): Contains the names of the datasets, i.e., coco_2014_trian, coco_2014_val, etc tr

Constructs a new box head. By default, uses ROIBoxHead, but if it turns out not to be enough, just register a new class and make it a par