str: a string describing the module

Filter images without ground truths.

Compute regression and classification targets for anchors in a single image. Args: flat_anchors (torch.Tensor): Multi-lev

Initialize layers of the head.

Merging patch bboxes into full image. Params: CLASSES (list): Label category. iou_thr (float): Threshold of IoU. version

Resize bounding boxes with ``results['scale_factor']``.

Set flag according to image aspect ratio. All set to 0.

Add extra files that are required to support MIM into the package. These files will be added by creating a symlink to the originals if the pa

Test function with test time augmentation.

Backward function.

Backward function.

Flip bboxes horizontally or vertically. Args: bboxes(ndarray): shape (..., 5*k) img_shape(tuple): (height, width)

Evaluate mAP of single image det result. Args: det_result (list[list]): [[cls1_det, cls2_det, ...], ...]. The outer list indi

Apply transformation `pred_bboxes` to `boxes`. Args: bboxes (torch.Tensor): Basic boxes. Shape (B, N, 4) or (N, 4) pre

Get box regression transformation deltas that can be used to transform the ``bboxes`` into the ``gt_bboxes``. Args: bboxes

Forward function.

Forward function.

Args: x (list[Tensor]): feature maps of multiple scales best_rbboxes (list[list[Tensor]]):

Forward features from the upstream network. Args: feats (tuple[Tensor]): Features from the upstream network, each is

Used for computing network flops. See `mmedetection/tools/get_flops.py`

Forward feature of a single scale level. Args: x (torch.Tensor): Features of a single scale level. Returns:

Forward feature of a single scale level. Args: x (torch.Tensor): Features of a single scale level. Returns:

Forward function.

Get anchors according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. img_met

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

Get the best begin point of the single polygon. Args: coordinate (List): [x1, y1, x2, y2, x3, y3, x4, y4, score] Returns: reo

Convert horizontal bounding boxes to oriented bounding boxes. Args: hbbs (torch.Tensor): [x_lt,y_lt,x_rb,y_rb] version (Str): ang

Initialize weights of the head.

Randomly decide whether to rotate.

Params: ann_folder: folder that contains DOTA v1 annotations txt files

Loss function of RRetinaRefineHead.

Compute loss of a single scale level. Args: cls_score (torch.Tensor): Box scores for each scale level Has shape (

list[int]: total number of base anchors in a feature grid

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

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

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. (r3det version) Args: polys (torch.Tensor): [x0,y0,x1,y1,x2,y2,x3,y3] version (S

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

Refine predicted bounding boxes at each position of the feature maps. This method will be used in R3Det in refinement stages. Args: