Project Setup

This project uses uv for fast, reproducible Python environments and dependency resolution.

📦 Install uv

On Windows (PowerShell)

curl -sSf https://astral.sh/uv/install.ps1 | iex

On Linux / macOS (bash/zsh):

curl -LsSf https://astral.sh/uv/install.sh | sh

ℹ️ After installation, make sure uv is on your PATH. You may need to restart your terminal or manually add ~/.cargo/bin to your shell config.

🔧 Environment Setup

Create and activate a virtual environment, then install dependencies with CUDA-enabled PyTorch:

On Windows (PowerShell)

uv venv .venv
.venv/Scripts/activate
uv sync

On Linux / macOS (bash/zsh):

uv venv .venv
source .venv/bin/activate
uv sync

Get Started

1. Download and clean the data set.

2. Slice the 3D Point Clouds(optionally visualize)

Default location of Point Clouds: ./data/raw/PointClouds/

python -m src.main slice

3. Prepare Dataset

# For prepaing without padding and masking:
python -m src.main prep

# For preparing with padding and masking:
python -m src.main prep --pad --target-points 6500

4. Train

python -m src.main train --config "path/to/config.json" --resume --fit-scalar

5. Evaluate
6. Predict