Return a new batched Transform3d representing the batch elements from self and all the given other transforms all batched together.

Create a 1D, 2D, or 3D convolution module.

Return a copy of this pose object. Returns ---------- result: Pose Copied pose object.

Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.

Extract features from images and points.

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

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

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

Return a matrix which is the result of composing this transform with others stored in self.transforms. Where necessary transforms

Return a new Transform3d representing the composition of self with the given other transforms, which will be stored as an internal li

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

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

Match functionality of torch.Tensor.to() If copy = True or the self Tensor is on a different device, the returned tensor is a

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

Deep copy of Transforms object. All internal tensors are cloned individually. Returns: new Transforms object.

Convert action.

Extract features from images and points.

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

Default initialization for Parameters of Module.

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

Obtain history BEV features iteratively. To save GPU memory, gradients are not calculated.

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

Adapted from AdelaiDet: https://github.com/aim-uofa/AdelaiDet/blob/master/adet/utils/comm.py

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

Zero out the parameters of a module and return it.

A Transform3d representing a scaling operation, with different scale factors along each coordinate axis. Option I: Scale(s,

Batch multiply two matrices and broadcast if necessary. Args: a: torch tensor of shape (P, K) or (M, P, K) b: torch tensor o

Performs the actual evaluation. :return: A tuple of high-level and the raw metric data.

obtain bev features.

Gets the device of the specified variable x if it is a tensor, or falls back to a default CPU device otherwise. Allows overriding by prov

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

Create a linear module.

Makes an actual torch.device object from the device specified as either a string or torch.device object. If the device is `cuda` without

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

Obtain history BEV features iteratively. To save GPU memory, gradients are not calculated.

Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion process from the learned m

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

Parameters ---------- points2d: Tensor xy coordinates. shape=(N, ..., 2) E.g., (N, 2) or (N, K, 2) or (N, H, W, 2) i

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

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

Velocity of ego-vehicle in m/s.

Helper function for _handle_input. Args: c: Python scalar, torch scalar, or 1D torch tensor Returns: c_vec: 1D torch te

Helper function to handle parsing logic for building transforms. The output is always a tensor of shape (N, 3), but there are several types o

This function is used to transpose image first training targets to level first ones :return: level first training targets

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

1x1 convolution with padding

3x3 convolution with padding

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

Encode bitmap mask to RLE code. Semantic Masks only Args: mask_results (list | tuple[list]): bitmap mask results. In mask scor

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

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 json (standard format for COCO evaluation). Args: results (list[tuple | numpy.ndarray]): Testing results of

img_feats (list[Tensor]): 5-D tensor for each level, (B, N, C, H, W) gt_bboxes (list[list[Tensor]]): Ground truth bboxes for each ima

Forward function' Args: pts_feats (list[torch.Tensor]): Features of point cloud branch gt_bboxes_3d (list[:obj:`BaseIn