Interpolate y, z values at each anchor grid, including those beyond the range of input lnae x range :param input_lane: N x 2 or N x 3 nda

Parse and save the images in jpg format. Jpg is the original format used by Waymo Open dataset. Saving in png format will cause huge (~3x)

Generate grid Points of a single level. Note: This function is usually called by method ``self.grid_priors``. Args:

Convert the model into training mode while keep normalization layer freezed.

A function wrapper for launching model training according to cfg. Because we need different eval_hook in runner. Should be deprecated in the

Process vertices.

Solving an Assignment Problem with MinCostFlow" :param adj_mat: adjacency matrix with binary values indicating possible matchings between

r""" Calculate AP on given distance threshold. Parameters ---------- gts : dict Dict storing ground truth for all samples.

r""" Calculate average precision on the given adjacent matrixs, where vertices are directedly connected. Parameters ---------- gt

r""" Calculate average precision on the given adjacent matrixs, where vertices are undirectedly connected. Parameters ----------

Args: re_zero: If True, alpha scale residuals with zero init.

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

Averages the input gradient tensor by allreduce and returns a future.

Run forward function and calculate loss for auxiliary head in training.

Run forward function and calculate loss for auxiliary head in training.

Args: cls_pred (Tensor): The prediction with shape (num_query, 1, *) or (num_query, *). gt_labels (Te

Convert class_id in loaded segmentation mask to label.

Convert class_id in loaded segmentation mask to label.

Run forward function and calculate loss for decode head in inference.

Run forward function and calculate loss for decode head in inference.

Run forward function and calculate loss for decode head in training.

Run forward function and calculate loss for decode head in training.

Convert detection results to COCO json style.

Embeds face sequences and adds within and between face positions. Args: seq: B, seqlen=vlen*3, condition_embedding:

Expand onehot labels to match the size of prediction.

Generate the train/val infos from the raw data. Args: nusc (:obj:`NuScenes`): Dataset class in the nuScenes dataset. train_scenes

Generate the train/val infos from the raw data. Args: nusc (:obj:`NuScenes`): Dataset class in the nuScenes dataset. train_scenes

Generate the train/val infos from the raw data. Args: lyft (:obj:`LyftDataset`): Dataset class in the Lyft dataset. train_scenes

Generate the train/val infos from the raw data. Args: nusc (:obj:`NuScenes`): Dataset class in the nuScenes dataset. train_scenes

Transform outputs from the last decoder layer into bbox predictions for each image. Args: cls_score (Tensor): Box score l

Return a process group based on gloo backend, containing all the ranks The result is cached.