Print out the loss function for current iteration. Parameters ---------- epoch : int Current epoch for

Parameters ---------- boxes: np.ndarray (N, 7) [x, y, z, dx, dy, dz, heading], (x, y, z) is the box center limit_range: list

r""" Normalize a given homography in pixels to [-1, 1]. Args: dst_pix_trans_src_pix : torch.Tensor Homography/ies from sou

Normalize label_map Parameters ---------- label_map : numpy.array Targets array for classification and r

Process the outputs of the model to 2D bounding box for debug purpose. Step1: convert each cav's output to bounding box format

Project points to BEV occupancy map with default ratio=0.1. Parameters ---------- points : np.ndarray (N

Reform the data params with current timestamp object groundtruth and delay timestamp LiDAR pose. Parameters --------

Regroup the data based on the record_len. Parameters ---------- dense_feature : torch.Tensor N, C, H, W record_len : lis

PyTorch Implementation for tf.nn.sigmoid_cross_entropy_with_logits: max(x, 0) - x * z + log(1 + exp(-abs(x))) in https://www.

Calculate the time delay for a certain vehicle. Parameters ---------- ego_flag : boolean Whether the cur

Update label_map based on bbx and regression targets. Parameters ---------- label_map : numpy.array Targ

Visualize the BEV 2D prediction, ground truth with point cloud together. Parameters ---------- pred_box_tensor : tor

Visualize a single frame of a single CAV for validation of data pipeline. Parameters ---------- o3d_pcd : o3d.PointCloud Ope

VOC 2010 Average Precision.

Abstract method, needs to be define by the children class.

Initialize the ConvLSTM cell :param input_size: (int, int) Height and width of input tensor as (height, width). :

Args: beta: Scalar float. L1 to L2 change point. For beta values < 1e-5, L1 loss is computed.

Convert the torch tensor bounding box to o3d aabb for visualization. Parameters ---------- bbx_center : torch.Tensor shape:

Convert 3D bounding box to 2D. Parameters ---------- box3d : np.ndarray (n, 8, 3) Returns ------- box2d : np.nd

0 -------- 1 | | | | | | 3 -------- 2 Parameters __________ boxes2d: np.ndarray

0 -------- 1 | | | | | | 3 -------- 2 Parameters __________ boxes3d: np.ndarray

Calculate the true positive and false positive numbers of the current frames. Parameters ---------- det_boxes : torch.Tensor

Customized collate function for target label generation. Parameters ---------- label_batch_list : list T

Customized collate function for target label generation. Parameters ---------- label_batch_list : list T

Customized pytorch data loader collate function. Parameters ---------- batch : list or dict List or dict

Customized pytorch data loader collate function. Parameters ---------- batch : list or dict List or dict

Customized collate function for pytorch dataloader during testing for late fusion dataset. Parameters ----------

Customized collate function for pytorch dataloader during testing for late fusion dataset. Parameters ----------

Customized collate function for pytorch dataloader during training for late fusion dataset. Parameters ----------

Customized collate function for pytorch dataloader during training for late fusion dataset. Parameters ----------

Compute iou between box and boxes list Parameters ---------- box : shapely.geometry.Polygon Bounding box Polygon. boxes

Convert boxes array to shapely.geometry.Polygon format. Parameters ---------- boxes_array : np.ndarray (N, 4, 2) or (N, 8, 3)

Find the minmaxx, minmaxy for each 2d box. (N, 4, 2) -> (N, 4) x1, y1, x2, y2 Parameters ---------- box2d : np.ndarray (

Find the minmax x and y for each bounding box. Parameters ---------- box_corner : torch.Tensor Shape: (N, 8, 3) or (N, 4)

Create the loss function based on the given loss name. Parameters ---------- hypes : dict Configuration params for training.