mdlearn.nn.modules.conv1d_encoder

Conv1dEncoder module for point cloud data.

Classes

Conv1dEncoder(*args, **kwargs)

class mdlearn.nn.modules.conv1d_encoder.Conv1dEncoder(*args: Any, **kwargs: Any)
__init__(num_points: int, num_features: int = 0, latent_dim: int = 20, bias: bool = True, relu_slope: float = 0.0, filters: List[int] = [64, 128, 256, 256, 512], kernels: List[int] = [5, 5, 3, 1, 1])

Conv1dEncoder module for point cloud data.

Parameters

  • num_points (int) – Number of input points in point cloud.

  • num_features (int, optional) – Number of scalar features per point in addition to 3D coordinates, by default 0.

  • latent_dim (int, optional) – Latent dimension of the encoder, by default 20.

  • bias (bool, optional) – Use a bias term in the Conv1d layers, by default True.

  • relu_slope (float, optional) – If greater than 0.0, will use LeakyReLU activiation with negative_slope set to relu_slope, by default 0.0.

  • filters (List[int], optional) – Encoder Conv1d filter sizes, by default [64, 128, 256, 256, 512].

  • kernels (List[int], optional) – Encoder Conv1d kernel sizes, by default [5, 5, 3, 1, 1].

forward(x: torch.Tensor) → Tuple[torch.Tensor, torch.Tensor]