Args: num_clusters : int The number of clusters dim : int Dimension of descriptors

:param ctx: :param grad_out: (B, C, N) tensor with gradients of outputs :return: grad_features: (B, C, M) tensor

:param ctx: :param grad_out: (B, C, npoint, nsample) tensor of the gradients of the output from forward :return:

Args: ctx: grad_out: (M1 + M2 ..., C, nsample) tensor of the gradients of the output from forward Returns:

Args: ctx: grad_out: (N1 + N2 ..., C) Returns: grad_features: (M1 + M2 ..., C)

Args: ctx: grad_new_features: (M1 + M2 ..., num_c_out), num_c_out = num_c_out_each_grid * num_total_grids Re

Converts depth map into bin indices Args: depth_map: (H, W), Depth Map mode: string, Discretiziation mode (See https://arxiv.

Builds frustum to voxel transformation Returns: f2v_module: nn.Module, Frustum to voxel transformation

Builds frustum feature network Returns: ffn_module: nn.Module, Frustum feature network

Performs all_gather operation on the provided tensors. *** Warning ***: torch.distributed.all_gather has no gradient.

:param corners3d: (N, 8, 3) corners in rect coordinate :return: boxes: (None, 4) [x1, y1, x2, y2] in rgb coordinate :return: