Initialization before data preparation. Args: results (dict): Dict before data preprocessing. - img_fields (lis

Get annotation info according to the given index. Args: index (int): Index of the annotation data to get. Returns:

Default initialization for Parameters of Module.

Create reduced point clouds for given info. Args: data_path (str): Path of original data. info_path (str): Path of data info.

nanmean compatible with generators.

Results visualization. Args: results (list[dict]): List of bounding boxes results. out_dir (str): Output directory

Decode bboxes. Args: all_cls_scores (Tensor): Outputs from the classification head, \ shape [nb_dec, bs, num_query

Args: input_ch(int) : the number of input channel out_features (list[str]): name of the layers whose outputs should

Build PyTorch DataLoader. In distributed training, each GPU/process has a dataloader. In non-distributed training, there is only one dataloade

Flattens predictions in the batch

Provides the default colors based on the category names. This method works for the general nuScenes categories, as well as the nuScenes det

Determine the (x,y,z) locations (in the ego frame) of the points in the point cloud. Returns B x N x D x H/downsample x W/downsample x

Loss function. Args: voxel_semantics (torch.Tensor): Shape (bs, w, h, total_z) valid_mask (torch.Tensor)

Obtain the info with RT matric from general sensor to Top LiDAR. Args: nusc (class): Dataset class in the nuScenes dataset. senso

Training data preparation. Args: index (int): Index for accessing the target data. Returns: dict: Tra

Renders various PR and TP curves. :param metrics: DetectionMetrics instance. :param md_list: DetectionMetricDataList instance

Get item from infos according to the given index. Args: idx (int): Index for accessing the target data. Returns:

The main evalution function. Called by function `eval` in dataset class. Args: CDist_tensor (np.array): (num_classes-1

Evaluation. Args: voxel_semantics (torch.Tensor): semantic occpuancy ground truth. voxel_semantics_preds

Extract features from images and points.

Extract features from images and points.

Extract features from images and points.

Flattens predictions in the batch (binary case) Remove labels equal to 'ignore'

Convert instance masks to axis-aligned bounding boxes. Args: points (np.array): Scene points of shape (n, 6). pts_ins

This method is used to obtain Bird's Eye View (BEV) features from multi-level features, BEV queries, and other related parameters.

get the data info according to the index. Most of them are image meta data. Args: index (Int): the index of the data.

Get data info according to the given index. Args: index (int): Index of the sample data to get. Returns:

Get data infos. This method gets information from the raw data. Args: num_workers (int): Number of threads to be used. D

KITTI annotation format version 2: { [optional]points: [N, 3+] point cloud [optional, for kitti]image: { image_id

Generate bboxes from bbox head predictions. Args: preds_dicts (tuple[list[dict]]): Prediction results. Return

Prepare data related to nuScenes dataset. Related data consists of '.pkl' files recording basic infos, 2D annotations and groundtruth databas

compute and print the final results. Called by function `eval` in dataset class. Args: None. All information are stor

Evaluation for a single model in nuScenes protocol. Args: result_path (str): Path of the result file. logger (loggi

Convert the results to the standard format. Args: results (list[dict]): Testing results of the dataset. jsonfile_pr

Convert results to kitti format for evaluation and test submission. Args: net_outputs (List[np.ndarray]): list of array storing t

Convert the box from camera to global coordinate. Args: info (dict): Info for a specific sample data, including the calibra

1x1 convolution with padding

3x3 convolution with padding

Convert action.

Convert range images to point cloud. Args: frame (:obj:`Frame`): Open dataset frame. range_images (dict): Mapping fro

Given the raw data, generate the ground truth database. Args: dataset_class_name （str): Name of the input dataset. data_path (str

Test model with multiple gpus. This method tests model with multiple gpus and collects the results under two different modes: gpu and cpu mo

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

Applies filtering to boxes. basedon overlap . :param nusc: An instance of the NuScenes class. :param eval_boxes: An instance of the EvalB

Format the results to pkl file. Args: outputs (list[dict]): Testing results of the dataset. pklfile_prefix (str | Non

Forward function Args: multi_level_feats (list[torch.Tensor]): Multi level img_feats. voxel_semantics (to

Forward inference function. Args: (all arg are be wrapped one more list. after we take it out, the type of each param

Forward inference function. Args: (all arg are be wrapped one more list. after we take it out, the type of each param

During inference, the functionary of batch norm layers is turned off but only the mean and var alone channels are used, which exposes the chance t

Get the 2D annotation records for a given info. Args: info: Information of the given sample data. occluded: Integer (0, 1, 2, 3)

Get available scenes from the input nuscenes class. Given the raw data, get the information of available scenes for further info generation.

Calculate the locations of the frustum points in the lidar coordinate system. Args: rots (torch.Tensor): Rotation from ca

Calculate multiple losses using different loss functions. Args: loss_occ_fun (dict): A dictionary containing different

Get the reference points used in SCA and TSA. Args: bev_z, bev_h, bev_w: spatial shape of voxel. D: sample D points

Get the reference points used in SCA and TSA. Args: H, W: spatial shape of bev. Z: hight of pillar. D:

Get the reference points used in SCA and TSA. Args: bev_z, bev_h, bev_w: spatial shape of voxel. D: sample D points

Get the reference points used in SCA and TSA. Args: H, W: spatial shape of bev. Z: hight of pillar. only

Get the reference points used in SCA and TSA. Args: H, W: spatial shape of bev. Z: hight of pillar. D:

Get the reference points used in SCA and TSA. Args: H, W: spatial shape of bev. Z: hight of pillar. only

obtain bev features.

Encoder for image features. Args: img (torch.Tensor): Image tensor. shape (B, N, C=3, H, W) Returns:

Initialize weights of the DeformDETR head.

Inverse function of sigmoid. Args: x (Tensor): The tensor to do the inverse. eps (float): EPS avoid numerical

Computes gradient of the Lovasz extension w.r.t sorted errors See Alg. 1 in paper