A masked version of nn.BatchNorm1d. Only tested for 3D inputs. Args: num_features: :math:`C` from an expected input of size

A residual block that can optionally change the number of channels. :param channels: the number of input channels. :param emb_channels:

A sequential module that passes timestep embeddings to the children that support it as an extra input.

An attention block that allows spatial positions to attend to each other. Originally ported from here, but adapted to the N-d case. http

Occupancy Network class. Args: decoder (nn.Module): decoder network encoder (nn.Module): encoder network device (device)

A downsampling layer with an optional convolution. :param channels: channels in the inputs and outputs. :param use_conv: a bool determin

`Large-scale CelebFaces Attributes (CelebA) Dataset <http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html>`_ Dataset. Args: root (string): Ro

An upsampling layer with an optional convolution. :param channels: channels in the inputs and outputs. :param use_conv: a bool determini

Holds a binvox model. data is either a three-dimensional numpy boolean array (dense representation) or a two-dimensional numpy float array (c

Crops the given PIL Image on the long edge. Args: size (sequence or int): Desired output size of the crop. If size is an int i

Decoder. Instead of conditioning on global features, on plane/volume local features. Args: dim (int): input dimension c_

PointNet-based encoder network with ResNet blocks for each point. Number of input points are fixed. Args: c_dim (int): dimension

A module which performs QKV attention and splits in a different order.

Fully connected ResNet Block class. Args: size_in (int): input dimension size_out (int): output dimension size_h (int):

Adapted from CLIP: https://github.com/openai/CLIP/blob/main/clip/model.py

A helper Module that performs 2 convolutions and 1 MaxPool. A ReLU activation follows each convolution.

The half UNet model with attention and timestep embedding. For usage, see UNet.

Pretrained InceptionV3 network returning feature maps

`LSUN <https://www.yf.io/p/lsun>`_ dataset. Args: root (string): Root directory for the database files. classes (string or l

Mesh evaluation class. It handles the mesh evaluation process. Args: n_points (int): number of points to be used for evaluation

A module which performs QKV attention. Matches legacy QKVAttention + input/ouput heads shaping

A UNetModel that performs super-resolution. Expects an extra kwarg `low_res` to condition on a low-resolution image.

Dumps key/value pairs into TensorBoard's numeric format.

`UNet` class is based on https://arxiv.org/abs/1505.04597 The U-Net is a convolutional encoder-decoder neural network. Contextual spatial in

The full UNet model with attention and timestep embedding. :param in_channels: channels in the input Tensor. :param model_channels: base

A helper Module that performs 2 convolutions and 1 UpConvolution. A ReLU activation follows each convolution.

Positional Encoding (presented in NeRF) Args: basis_function (str): basis function

A class replacing the SVD of a matrix A, perhaps efficiently. All input vectors are of shape (Batch, ...). All output vectors are of shap

Global Variables & Strctures

FKAConv convolution layer. To be used with a `lightconvpoint.nn.Conv` instance. # Arguments in_channels: int. The number

FKAConv convolution layer. To be used with a `lightconvpoint.nn.Conv` instance. # Arguments in_channels: int. The number

ConvPoint convolution layer. Provide the convolution layer as defined in ConvPoint paper (https://github.com/aboulch/ConvPoint). To be us