π Documentationβ’ π Quickstart β’ β¨ Features β’ βοΈ Installation β’ π License
πPADO (νλ) is a cutting-edge framework for differentiable optical simulations powered by PyTorch. Inspired by the Korean word for "wave," PADO enables seamless and fully differentiable simulation workflows, perfect for researchers and developers in optical physics, computational imaging, and beyond.
- π₯ Fully Differentiable: Integrates effortlessly with PyTorch Autograd.
- ποΈ CUDA Acceleration: Leverages GPU hardware for ultra-fast simulations.
- 𧩠Modular Components: Easily customizable optical elements and simulation environments.
- π Visualization Tools: Rich visualization with Matplotlib.
- β‘ < 8000 strong>Easy-to-use API: Beginner-friendly API for rapid experimentation.
You can install PADO via pip:
pip install pado-opticsOr via conda:
conda install -c conda-forge pado-opticsOr install directly from GitHub:
pip install git+https://github.com/shwbaek/pado.gitFor development installation:
git clone https://github.com/shwbaek/pado.git
cd pado
pip install -e .Comprehensive documentation is available at https://shwbaek.github.io/pado.
PADO includes a comprehensive set of example notebooks organized by topic:
Browse our examples by category:
-
- 1.1 Pado fundamentals - Learn about core components and building blocks
- 1.2 RGB multi-wavelength - Working with multiple wavelengths
- 1.3 4-F with batch - Batch processing in 4-F systems
- 1.4 How to use ASM options - Angular Spectrum Method configuration
-
2. Computer Generated Holography
- 2.1 DPAC - Double Phase Amplitude Coding
- 2.2 Multi-depth CGH - Multi-plane holography
- 2.3 CGH optimization - GS, SGD, and Adam methods
- 2.4 Multi-depth hologram with Adam - Complex loss-based optimization
- 2.5 Phase-only SLM optimization - Optimization with phase-only spatial light modulators
- 2.6 Multi-depth hologram with phase-only SLM - Multi-plane optimization with phase-only SLMs
-
- 3.1 Lens comparison - Different lens models and wavefront observation
- 3.2 Coded aperture comparison - Coded aperture techniques
- 3.3 Seeing through DOE - Imaging through diffractive optical elements
-
- 4.1 Polarization light - Polarized light simulation
-
- 5.1 Chromatic aberration singlet - Chromatic aberration simulation
Developed and maintained by the POSTECH Computer Graphics Lab.
This project is licensed under the MIT License - see the LICENSE file for details.
If you use Pado in your research, please cite Pado using the following BibText template:
@misc{Pado,
Author = {Seung-Hwan Baek, Dong-Ha Shin, Yujin Jeon, Seung-Woo Yoon, Eunsue Choi, Gawoon Ban, Hyunmo Kang},
Year = {2025},
Note = {https://github.com/shwbaek/pado},
Title = {Pado: Pytorch Automatic Differentiable Optics}
}