fastdev.nn.spconv_unet

SparseUNet Driven by SpConv.

Adapted from: https://github.com/Pointcept/Pointcept

This module requires the installation of the following packages:

• spconv: https://github.com/traveller59/spconv?tab=readme-ov-file#spconv-spatially-sparse-convolution-library

Original Author: Xiaoyang Wu (xiaoyang.wu.cs@gmail.com) Please cite their work if you use the following code in your research paper.

Module Contents

class fastdev.nn.spconv_unet.SpUNetBase(in_channels: int, num_classes: int, base_channels=32, channels=(32, 64, 128, 256, 256, 128, 96, 96), layers=(2, 3, 4, 6, 2, 2, 2, 2), cls_mode=False)

Bases: torch.nn.Module

Parameters:

• in_channels (int)

• num_classes (int)

in_channels

num_classes

base_channels = 32

channels = (32, 64, 128, 256, 256, 128, 96, 96)

layers = (2, 3, 4, 6, 2, 2, 2, 2)

num_stages

cls_mode = False

conv_input

down

up

enc

dec

final

forward(points: torch.Tensor)

Parameters:

points (torch.Tensor)

class fastdev.nn.spconv_unet.SpUNetCls(in_channels: int, num_classes: int, base_channels=32, channels=(32, 64, 128, 256, 256, 128, 96, 96), layers=(2, 3, 4, 6, 2, 2, 2, 2))

Bases: torch.nn.Module

Parameters:

• in_channels (int)

• num_classes (int)

model

forward(points: torch.Tensor)

Parameters:

points (torch.Tensor)

class fastdev.nn.spconv_unet.SpUNetSeg(in_channels: int, num_classes: int, base_channels=32, channels=(32, 64, 128, 256, 256, 128, 96, 96), layers=(2, 3, 4, 6, 2, 2, 2, 2))

Bases: torch.nn.Module

Parameters:

• in_channels (int)

• num_classes (int)

model

forward(points: torch.Tensor)

Parameters:

points (torch.Tensor)