Circular Smooth Label Decoder. Args: angle_preds (Tensor): The csl encoding of angle offset for each scale level.

Digit version.

Build loss.

Assign gt to bboxes. The assignment is done in following steps 1. compute iou between all bbox (bbox of all pyramid levels) and gt

Limit the range of angles. Args: angle (ndarray): shape(n, ). angle_range (Str): angle representations. Returns: ang

build a enn regular feature map with the specified number of channels divided by N.

Circular Smooth Label Encoder. Args: angle_targets (Tensor): Angle offset for each scale level Has shape (num_anc

Sample positive and negative bboxes. This is a simple implementation of bbox sampling given candidates, assigning results and ground

Loss function of ODMRefineHead.

Builder of box assigner.

Builder of box coder.

Build head.

NMS for multi-class bboxes. Args: multi_bboxes (torch.Tensor): shape (n, #class*5) or (n, 5) multi_scores (torch.Tensor): shape (

build an enn normalizion layer.

enn ReLU.

Convert oriented bounding boxes to polygons. Args: obbs (torch.Tensor): [x_ctr,y_ctr,w,h,angle] version (Str): angle representati

Deprecated sample_per_gpu in cfg.data.

enn convolution. Args: in_channels (List[int]): Number of input channels per scale. out_channels (int): Number of output channels

Concat multi-level feature maps by image. [feature_level0, feature_level1...] -> [feature_image0, feature_image1...] Convert the shape of eac

Convert oriented bounding boxes to horizontal bounding boxes. Args: obbs (torch.Tensor): [x_ctr,y_ctr,w,h,angle] version (Str): a

Convert polygons to oriented bounding boxes. Args: polys (ndarray): [x0,y0,x1,y1,x2,y2,x3,y3] version (Str): angle representation

Convert detection results to a list of numpy arrays. Args: bboxes (torch.Tensor): shape (n, 6) labels (torch.Tensor): shape (n, )

Box head forward function used in both training and testing. Args: x (list[Tensor]): list of multi-level img features.

Build detector.

Builder of box sampler.

Forward function.

Kullback-Leibler Divergence loss. Args: pred (torch.Tensor): Predicted bboxes. target (torch.Tensor): Corresponding gt bboxes.

Get normalizion layer's name.

nn.Module: normalization layer after the first convolution layer

nn.Module: normalization layer after the second convolution layer

Parse the package dependencies listed in a requirements file but strips specific versioning information. Args: fname (str): path to r

Convert polygons to oriented bounding boxes. Args: polys (torch.Tensor): [x0,y0,x1,y1,x2,y2,x3,y3] version (Str): angle represent

This function will be used in S2ANet, whose num_anchors=1.

Setup multi-processing environment variables.

Transform outputs for a single batch item into bbox predictions. Args: cls_scores (list[Tensor]): Box scores of all scale level

Evaluate mAP of a rotated dataset. Args: det_results (list[list]): [[cls1_det, cls2_det, ...], ...]. The outer list indicates

Directly extract features from the backbone+neck.

Convert oriented bounding boxes to polygons. Args: obbs (ndarray): [x_ctr,y_ctr,w,h,angle] version (Str): angle representations.

Convert distance to loss. Args: distance (torch.Tensor) fun (str, optional): The function applied to distance. Defaul

Convert a list of bboxes to roi format. Args: bbox_list (list[Tensor]): a list of bboxes corresponding to a batch of images.

Build backbone.

Build neck.

Calculate the length of line. Args: point1 (List): [x,y] point2 (List): [x,y] Returns: length (float)

Make sure that the shape of x is (T, n, 1, d). Args: x (torch.Tensor): input tensor. Returns: torch.Tensor:

Construct toy data.

Directly extract features from the backbone+neck.

Transform network output for a batch into bbox predictions. Args: rois (torch.Tensor): Boxes to be transformed. Has shape

Calculate the ground truth for all samples in a batch according to the sampling_results. Almost the same as the implementation in bbo

Loss function. Args: cls_score (torch.Tensor): Box scores, has shape (num_boxes, num_classes + 1). bb

Make divisible function. This function rounds the channel number to the nearest value that can be divisible by the divisor. It is taken from

Convert the request input to a ndarray. Args : data (list): List of the data from the request input. Returns:

Calculate overlap between two set of bboxes. Args: bboxes1 (torch.Tensor): shape (B, m, 5) in <cx, cy, w, h, a> format or emp

Check whether the rotated anchors are inside the border. Args: flat_anchors (torch.Tensor): Flatten anchors, shape (n, 5). valid_

Add shared or separable branch. convs -> avg pool (optional) -> fcs

Add shared or separable branch. convs -> avg pool (optional) -> fcs

Box head forward function used in both training and testing. Args: x (list[Tensor]): list of multi-level img features.

Box head forward function used in both training and testing. Args: x (list[Tensor]): list of multi-level img features.

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions n

Compute overlaps between polygons and points. Args: gt_rbboxes (torch.Tensor): Groundtruth polygons, shape (k, 8). po

Evaluate the dataset. Args: results (list): Testing results of the dataset. metric (str | list[str]): Metrics to be e

Directly extract features from the backbone+neck.

Directly extract features from the backbone+neck.

Filter predicted bounding boxes at each position of the feature maps. Only one bounding boxes with highest score will be left at each

Find the latest checkpoint from the working directory. Args: path(str): The path to find checkpoints. suffix(str): File extension

Transform network output for a batch into labeled boxes. Args: cls_scores (list[Tensor]): Box scores for each scale level

Get the best begin points of polygons. Args: coordinate (ndarray): shape(n, 9). Returns: reorder coordinate (ndarray): shape

Get palette from various inputs. Args: palette (list[tuple] | str | tuple | :obj:`Color`): palette inputs. num_classes (int): the

Compute regression and classification targets for anchors in multiple images. Args: anchor_list (list[list[Tensor]]): Mul

make plugins for block. Args: in_channels (int): Input channels of plugin. plugins (list[dict]): List of plugins cfg

Convert oriented bounding boxes to horizontal bounding boxes. Args: obbs (torch.Tensor): [x_ctr,y_ctr,w,h,angle] version (Str): a

Convert oriented bounding boxes to polygons. Args: obbs (torch.Tensor): [x_ctr,y_ctr,w,h,angle] Returns: polys (torch.Tensor

Sample edge points for polygon. Args: polygons (torch.tensor): polygons with shape (N, 8) points_num (int): number of

Compute the Chamfer distance between two point sets. Args: point_set_1 (torch.tensor): point set 1 with shape (N_pointsets,

Check data in fields from two results are same.

Freeze stages.

Get adaptive scales according to areas. The scale range is [0.5, 1.0]. When the area is less than ``'min_area'``, the scale is 0.5 while the

Transform outputs for a single batch item into bbox predictions. Args: cls_score_list (list[Tensor]): Box scores for a single sca

Transform outputs for a single batch item into bbox predictions. Args: cls_score_list (list[Tensor]): Box scores for a single sca

Grab configs necessary to create a detector. These are deep copied to allow for safe modification of parameters without influencing other tes

Initializes the parameters of Gaussian mixture model. Args: mu_init (torch.Tensor, optional): mu of Gaussian. var_ini

Replace ResNet50 with ResNet18 in config.

Backward function.

build a enn regular feature map with the specified number of channels.

build a enn trivial feature map with the specified number of channels.

Build roi extractor.

Compute centerness targets. Args: pos_bbox_targets (Tensor): BBox targets of positive bboxes in shape (num_pos, 4

Collect environment information.

This function would modify some filed to keep the compatibility of config. For example, it will move some args which will be deprecated to th