Installation¶

Creating a Virtual Environment¶

Before installing any dependencies, I recommend creating a new virtual environment to avoid library conflicts.

conda create -n deeplens python=3.11
conda activate deeplens

The library should work with Python 3.11+, but I recommend using Python 3.11 specifically, as it was used during development. This will help prevent version mismatches and dependency conflicts.

Standard Installation¶

To install DeepLens and its dependencies, run the following commands:

pip install deeplens-sae

# FOR WINDOWS
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu128
# OR FOR MAC
pip install torch torchvision

Alternative Installation¶

If you encounter any errors with the standard installation, try the installation from source:

git clone https://github.com/iparramartin/deeplens
cd deeplens

# FOR WINDOWS
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu128
# OR FOR MAC
pip install torch torchvision

pip install -e .

Common Version Issues¶

During installation, you may encounter version conflicts. The most common issues involve numpy and torch versions.

NumPy Version¶

We recommend installing numpy<2, as this version was used during development and has shown the fewest conflicts with other libraries (e.g., scipy).

PyTorch Installation¶

For CUDA support:

pip install torch torchvision --index-url https://download.pytorch.org/whl/cu128

For MPS/CPU usage:

pip install torch torchvision

Future Updates¶

PyPI installation is not yet available. Future versions will support easier installation via pip.

Ready to dive deeper? Check out the Quickstart Guide for complete examples!