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

Process one batch of data samples and predictions. The processed results should be stored in ``self.results``, which will be used to c

Exchange bbox_head key to rpn_head key when loading single-stage weights into two-stage model.

Add predictions to `DetDataSample`. Args: data_samples (list[:obj:`DetDataSample`], optional): A batch of data sa

Get annotation by index and automatically call ``full_init`` if the dataset has not been fully initialized. Args: idx (in

Predict results from a batch of inputs and data samples with post- processing. Args: batch_inputs (Tensor): Inputs with s

Calculate losses from a batch of inputs and data samples. Args: batch_inputs (Tensor): Input images of shape (N, C, H, W).

add new dimensions at the front and the tail

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

In distributed setting, this function only download checkpoint at local rank 0.

Transform a batch of output features extracted from the head into bbox results. Args: rois (tuple[Tensor]): Tuple of boxe

Forward function for feature maps before classifying each pixel with ``self.cls_seg`` fc. Args: inputs (list[Tensor]): Li

Get subset of data information list. Args: indices (int or Sequence[int]): If type of indices is int, indices rep

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

Transform a batch of output features extracted from the head into mask results. Args: mask_preds (tuple[Tensor]): Tuple o

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

DCNv3 Module :param channels :param kernel_size :param stride :param pad :param dilation :par

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

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

Assign gt to bboxes. This method assign a gt bbox to every bbox (proposal/anchor), each bbox will be assigned with -1, or a semi-posi

Perform forward propagation and loss calculation of the bbox head on the features of the upstream network. Args: x (tuple

Multiple with alpha and add beat with clip. Args: img (np.ndarray): The input image. alpha (int): Image weights, chan

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

Extract features. Args: batch_inputs (Tensor): Image tensor with shape (N, C, H ,W). Returns: tuple[Tensor]:

Extract features from images.

Extract features. Args: batch_inputs (Tensor): Image tensor with shape (N, C, H ,W). Returns: tuple[Tensor]:

Inference with slide/whole style. Args: inputs (Tensor): The input image of shape (N, 3, H, W). batch_img_metas (List

Calculate losses from a batch of inputs and data samples. Args: inputs (Tensor): Input images. data_samples (list[:ob

Calculate the loss based on the features extracted by the detection head. Args: cls_scores (list[Tensor]): Box scores for

Parses the raw outputs (losses) of the network. Args: losses (dict): Raw output of the network, which usually contain losses

Perform forward propagation of the bbox head and predict detection results on the features of the upstream network. Args:

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