WarpConvNet

Welcome to the WarpConvNet documentation.

WarpConvNet is a high-performance 3D deep learning library built on NVIDIA's Warp framework. It provides efficient implementations of point cloud processing, sparse voxel convolutions, attention mechanisms for 3D data, and geometric operations.

🚀 Quick Start

import warpconvnet as wcn

# Create geometry from point cloud
geometry = wcn.geometry.Points(coords, features)

# Voxelize for sparse convolution
voxels = geometry.voxelize(voxel_size=0.05)

# Apply sparse convolution
features = voxels.sparse_conv(kernel_size=3)

📚 Documentation Sections

  • Getting Started: Installation and quick start guides
  • User Guide: Comprehensive tutorials and concepts
  • API Reference: Complete API documentation
  • Examples: Real-world usage examples
  • Diagrams: Architecture and data flow visualizations