Skip to content

Installation

There are several ways to use DeepLens. Which one fits depends on whether you want to extend the optics engine or simply call it as a library.

DeepLens is a research framework, and the intended workflow is to design custom optical systems by writing scripts inside a clone of the repository — next to the bundled examples (0_hello_geolens.py, 1_design_geolens.py, …). You get the full source to read, extend, and modify, which is what you want when implementing new surfaces, lens models, or end-to-end pipelines.

git clone https://github.com/vccimaging/DeepLens.git
cd DeepLens
# create the environment, then write your own script in the repo root

When you run a script from the repository root, import deeplens resolves to the local deeplens/ package, so your edits take effect immediately with no reinstall.

Install from PyPI

To use DeepLens as a dependency in your own project, install the published deeplens-core package:

pip install deeplens-core

from deeplens import GeoLens

Use this when you want the stable API as a library and don't need to modify the source.

Install from source

Install the latest (unreleased) code directly from GitHub:

pip install git+https://github.com/vccimaging/DeepLens.git

Or clone and install in editable mode, which combines a clone with a proper package install so deeplens is importable from anywhere while your local edits still take effect:

git clone https://github.com/vccimaging/DeepLens.git
cd DeepLens
pip install -e .

Requirements

  • Python >= 3.12
  • PyTorch (install a CUDA build for GPU acceleration — see Setup)
  • A CUDA-capable GPU is recommended but not required

Verify the install

import torch
from deeplens import GeoLens

print(torch.cuda.is_available())  # True if a CUDA GPU is available