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

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

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

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

Forward features of a single scale level. Args: x (Tensor): FPN feature maps of the specified stride. scale (:obj: `m

Forward feature map of a single FPN level.

Forward feature map of a single FPN level.

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 feature map of a single scale level.

Forward feature map of a single FPN level. Args: x (torch.tensor): single-level feature map sizes. Returns: c

Args: x (list[Tensor]): list of multi-level img features. img_metas (list[dict]): list of image info dict where each

Args: x (list[Tensor]): list of multi-level img features. img_metas (list[dict]): list of image info dict where each

Args: x (list[Tensor]): list of multi-level img features. img_metas (list[dict]): list of image info dict where each

Args: img (Tensor): of shape (N, C, H, W) encoding input images. Typically these should be mean centered and std

Forward function.

Args: img (Tensor): Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

Forward function of S2ANet.

Convert Gaussian distribution to polygons by SVD. Args: gmm (dict[str, torch.Tensor]): Dict of Gaussian distribution. Returns:

Get the points features from the locations of predicted points. Args: features (torch.tensor): base feature with shape (B,C,W,H)

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

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

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. Args: cls_scores (list[Tensor]): Box scores for each scale level

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

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

Transform network outputs of a batch into bbox results. Args: cls_scores (list[Tensor]): Classification scores for all

Transform network outputs of a batch into bbox results. Args: cls_scores (list[Tensor]): Classification scores for all

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

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

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

Transform network outputs of a batch into bbox results. Args: cls_scores (list[Tensor]): Classification scores for all

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

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

Calculate loss of all potential positive samples obtained from first match process. Args: cls_score (Tensor): Box scores

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

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

Gaussian Wasserstein distance loss. Args: pred (torch.Tensor): Predicted bboxes. target (torch.Tensor): Corresponding gt bboxes.

Gaussian Wasserstein distance loss. Derivation and simplification: Given any positive-definite symmetrical 2*2 matrix Z: :math

Predict the results given input request. Args: data (list[ndarray]): The list of a ndarray which are ready to pro

Single initial stage loss function.

Initialize weights of the head.

Initialize weights of feature refine module.

Load the model.pt file and initialize the MMRotate model object. Args: context (context): JSON Object containing information

Randomly decide whether to rotate.

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

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

Symmetrical Max Kullback-Leibler Divergence loss. Args: pred (torch.Tensor): Predicted bboxes. target (torch.Tensor): Correspondi

Symmetrical Min Kullback-Leibler Divergence loss. Args: pred (torch.Tensor): Predicted bboxes. target (torch.Tensor): Correspondi

Load annotation from XML style ann_file. Args: ann_file (str): Path of Imageset file. Returns: list[dict]: A

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

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

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

Load annotation from XML style ann_file. Args: ann_file (str): Path of Imageset file. Returns: list[dict]: A

Loss function of RotatedRetinaRefineHead.

Compute loss of the head. Args: cls_scores (list[Tensor]): Box scores for each scale level, each is a 4D-tensor, t

Loss function of KFIoUODMRefineHead.

Compute loss of the head. Args: cls_scores (list[Tensor]): Box scores for each scale level, each is a 4D-tensor, t

Loss function of SAM RepPoints head.

Loss function of CFA head.

Compute losses of the head. Args: cls_scores (list[Tensor]): Box scores for each scale level Has shape (N, num_an

Compute losses of the head. Args: cls_scores (list[Tensor]): Box scores for each scale level Has shape (N, num_an

Compute losses of the head. Args: cls_scores (list[Tensor]): Box scores for each scale level Has shape (N, num_an

Loss function of KFIoURRetinaRefineHead.

Loss function of OrientedRepPoints head.

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

Loss function.

Single loss function.

Single loss function.

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

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

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

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

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

Plot curve.

Assess the quality of each point set from the classification, localization, orientation, and point-wise correlation based on the assig

Convert the output from the inference and converts into a Torchserve supported response output. Args: data (list[Tensor])

CFA reassign process. Args: pos_losses (Tensor): Losses of all positive samples in single image. labe

This function will be used in S2ANet, whose num_anchors=1. Args: cls_scores (list[Tensor]): Box scores for each scale level

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

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

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

Refine bboxes during training. Args: rois (torch.Tensor): Shape (n*bs, 5), where n is image number per GPU, and b

Draw `result` over `img`. Args: img (str or Tensor): The image to be displayed. result (Tensor or tuple): The results

Test without augmentation. Args: x (list[Tensor]): list of multi-level img features. proposal_list (list[Tensors]): l

Test without augmentation. Args: x (list[Tensor]): list of multi-level img features. proposal_list (list[Tensors]): l

Test without augmentation.

Test function without test time augmentation. Args: imgs (list[torch.Tensor]): List of multiple images img_metas (lis