Calculate the yaw angle from a quaternion. Note that this only works for a quaternion that represents a box in lidar or global coordinate fra

Flip 3D data randomly. Args: input_dict (dict): Result dict from loading pipeline. direction (str, optional): Flip di

Flip 3D data randomly. Args: input_dict (dict): Result dict from loading pipeline. direction (str): Flip direction. D

Randomly sample an img_scale when ``multiscale_mode=='range'``. Args: img_scales (list[tuple]): Images scale range for sampling.

This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch, num_key_value_heads, seqlen, he

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Rotated box iou running in gpu. 500x faster than cpu version (take 5ms in one example with numba.cuda code). convert from [this project]( http

Sampling all categories of bboxes. Args: gt_bboxes (np.ndarray): Ground truth bounding boxes. gt_labels (np.ndarray):

Args: sample_points: 3D sampling points in shape [B, Q, T, G, P, 3] mlvl_feats: list of multi-scale features from neck, each in s

Save checkpoint to file. The checkpoint will have 3 fields: ``meta``, ``state_dict`` and ``optimizer``. By default ``meta`` will contain versi

Parse and save the images in png format. Args: frame (:obj:`Frame`): Open dataset frame proto. file_idx (int): Curren

Return semantic segmentation predictions in the original resolution. The input images are often resized when entering semantic segmentor. Mo

Calculates the image embeddings for the provided image, allowing masks to be predicted with the 'predict' method. Arguments:

Set template array. Args: array (tuple | list | int | float | np.ndarray | torch.Tensor): Template array.

Turn a list of integer scalars or integer Tensor scalars into a vector, in a way that's both traceable and scriptable. In tracing, `x` s

Visualize the points cloud. Args: save_path (str, optional): path to save image. Default: None.

Loss used in RetinaNet for dense detection: https://arxiv.org/abs/1708.02002. Args: inputs: A float tensor of arbitrary shape.

Test function without augmentaiton.

Test det bboxes without test-time augmentation. Args: feats (tuple[torch.Tensor]): Multi-level features from the

Test without augmentation.

Obtain mask prediction without augmentation.

Test without augmentation, only for ``RPNHead`` and its variants, e.g., ``GARPNHead``, etc. Args: x (tuple[Tensor]): Feat

Return `num_samples` (or fewer, if not enough found) random samples from `labels` which is a mixture of positives & negatives. It will tr

Compute and display summary metrics for evaluation results given a custom value for max_dets_per_image

Args: see documentation of :func:`paste_masks_in_image`.

Convert InstancesList to List[Instances]. The input `instances_list` can also be a List[Instances], in this case this method is a non

Convert the model into training mode while keep layers freezed.

Convert the model into training mode while keep layers freezed.

each line with 1 meter width pred_lines: num_preds, npts, 2 gt_lines: num_gts, npts, 2

Args: results (OrderedDict[dict]): task_name -> {metric -> score} Returns: bool: whether the verification succeeds or not

Data preparation for voxel pooling. Args: coor (torch.tensor): Coordinate of points in the lidar space in shape (

Apply dynamic voxelization to points. Args: points (list[torch.Tensor]): Points of each sample. Returns: tup

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Partition into non-overlapping windows with padding if needed. Args: x (tensor): input tokens with [B, H, W, C]. window_size

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_siz

Window unpartition into original sequences and removing padding. Args: x (tensor): input tokens with [B * num_windows, window_size, w

Helper function to Accumulate() Considers the evaluation results applicable to a particular category, area range, and max_detections parameter setting

Helper function to Accumulate() Compute a precision recall curve given a sorted list of detected instances encoded in evaluations, evaluation_indices,

For each IOU threshold, greedily match each detected instance to a ground truth instance (if possible) and store the results

Sort detections from highest score to lowest, such that detection_instances[detection_sorted_indices[t]] >= detection_instances[detection_sorted_indic

Partition the ground truth objects based on whether or not to ignore them based on area

Call function to pad images, masks, semantic segmentation maps. Args: results (dict): Result dict from loading pipeline. R

Call function to load multiple types annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`. R

:param center: Center of box given as x, y, z. :param size: Size of box in width, length, height. :param orientation: Box ori

Args: model (nn.Module): model to apply EMA. decay (float): ema decay reate. updates (int): counter of EM

Initialize a DetectionEval object. :param nusc: A NuScenes object. :param config: A DetectionConfig object. :param re