Transform a batch of output features extracted by the head into bbox results. Args: cls_scores (list[Tensor]): Classificat

Extract features.

Extract features.

Extract features.

Get annotation by index.

Freeze the model.

Freeze the parameters of the specified stage so that they are no longer updated.

``full_init`` dataset.

Call function to collect indexes. Args: dataset (:obj:`Dataset` or list): The dataset or cached list. Returns:

Calculate losses from a batch of inputs and data samples.

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

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

Parse raw annotation to target format. Args: raw_data_info (dict): Raw data information load from ``ann_file`` Returns:

Predict results from a batch of inputs and data samples with post- processing.

Parse a version string into a tuple. Returns: tuple[int | str]: The version info, e.g., "1.3.0" is parsed into (1, 3, 0), and

Wrapper for `efficient_nms` with TensorRT. Args: boxes (Tensor): The bounding boxes of shape [N, num_boxes, 4]. scores (Tensor): T

Wrapper for `efficient_nms` with TensorRT. Args: boxes (Tensor): The bounding boxes of shape [N, num_boxes, 4]. scores (Tensor): T

Calculate global coordinate of mosaic image and local coordinate of cropped sub-image. Args: loc (str): Index for the sub

Calculate global coordinate of mosaic image. Args: loc (str): Index for the sub-image. img_shape_hw (Sequence[int]):

Update label text.

Add all parameters of module to the params list. The parameters of the given module will be added to the list of param groups, with s

Wrapper function for `_batched_nms`.

Mixed image data transformation. Args: results (dict): Result dict. Returns: results (dict): Updated result