r"""A sparse tensor class. Can be accessed via :attr:`MinkowskiEngine.SparseTensor`. The :attr:`SparseTensor` class is the basic tensor in Mi

r"""Convolution layer for a sparse tensor. .. math:: \mathbf{x}_\mathbf{u} = \sum_{\mathbf{i} \in \mathcal{N}^D(\mathbf{u}, K,

* Will be exported to python for lazy key initialization. For * instance, `sparse_tensor.coords_key` can be used for other layers * before feedforw

r"""A generalized sparse transposed convolution or deconvolution layer.

r"""A batch normalization layer for a sparse tensor. See the pytorch :attr:`torch.nn.BatchNorm1d` for more details.

A simple timer.

r"""Remove specified coordinates from a :attr:`MinkowskiEngine.SparseTensor`.

r"""Converts a (differentiable) dense tensor or a :attr:`MinkowskiEngine.TensorField` to a :attr:`MinkowskiEngine.SparseTensor`. For dense tensor

r"""Broadcast the reduced features to all input coordinates. .. math:: \mathbf{y}_\mathbf{u} = \mathbf{x}_{1, \mathbf{u}} + \mathbf{x}_2

r"""A sparse tensor class. Can be accessed via :attr:`MinkowskiEngine.SparseTensor`. The :attr:`SparseTensor` class is the basic tensor in Mi

r"""Broadcast reduced features to all input coordinates. .. math:: \mathbf{y}_\mathbf{u} = \mathbf{x}_{1, \mathbf{u}} \times \mathbf{x}_

r"""Max pool all input features to one output feature at the origin. .. math:: \mathbf{y} = \max_{\mathbf{i} \in \mathcal{C}^\te

r"""Sample linearly interpolated features at the provided points.

Samples elements randomly, without replacement. Arguments: data_source (Dataset): dataset to sample from

r"""Average input features within a kernel. .. math:: \mathbf{y}_\mathbf{u} = \frac{1}{|\mathcal{N}^D(\mathbf{u}, \mathcal{C}^\t

r"""Broadcast reduced features to all input coordinates and concatenate to the input. .. math:: \mathbf{y}_\mathbf{u} = [\mathbf{x}_{1,\

r"""A generalized sparse transposed convolution or deconvolution layer that generates new coordinates.

r"""A instance normalization layer for a sparse tensor.

r"""A max pooling layer for a sparse tensor. .. math:: y^c_\mathbf{u} = \max_{\mathbf{i} \in \mathcal{N}^D(\mathbf{u}, \mathcal{

r"""A pooling transpose layer for a sparse tensor. Unpool the features and divide it by the number of non zero elements that contributed.

r"""Sum all input features within a kernel. .. math:: \mathbf{y}_\mathbf{u} = \sum_{\mathbf{i} \in \mathcal{N}^D(\mathbf{u}, \ma

r"""Create a union of all sparse tensors and add overlapping features. Args: None .. warning:: This function is experimental

Multi-head attention to increase model expressivitiy

self attention block should encode: (1)local feature pattern from b*f*m*1+b*f*m*k (2)orientation consistance from b*g

* @brief Indicates whether a collection of values is known to be sorted. */

Samples elements randomly, without replacement. Arguments: data_source (Dataset): dataset to sample from

r"""Broadcast reduced features to all input coordinates. .. math:: \mathbf{y}_\mathbf{u} = \mathbf{x}_2 \; \text{for} \; \mathbf{u} \in

r""" Segmentation Module for MinkowskiEngine.

r"""A batch normalization layer with multi GPU synchronization.

r"""Converts a (differentiable) sparse tensor to a torch tensor. The return type has the BxCxD1xD2x....xDN format. Example:: >>> den

* @brief Exception thrown when a CUDA error is encountered. * * @ingroup utility_error */