Installation

DeepForest has Windows, Linux, and macOS prebuilt wheels on PyPI. We strongly recommend using uv or a virtualenv to create a clean installation container.

pip install deepforest

uv add deepforest

Source Installation

DeepForest can alternatively be installed from source using the GitHub repository. This can be done directly using pip or uv.

pip install git+https://github.com/weecology/DeepForest.git

uv add "deepforest @ git+https://github.com/weecology/deepforest"

Or you can clone and install locally.

git clone https://github.com/weecology/DeepForest.git
cd DeepForest
pip install .

git clone https://github.com/weecology/DeepForest.git
cd DeepForest
uv sync --all-extras --dev

Development Installation

For developers who want to contribute to DeepForest, install the package in development mode with all dependencies:

git clone https://github.com/weecology/DeepForest.git
cd DeepForest
pip install .'[dev,docs]'

Or using uv:

git clone https://github.com/weecology/DeepForest.git
cd DeepForest
uv sync --all-extras --dev

This installs DeepForest in editable mode with development and documentation dependencies.

GPU support

PyTorch can be run on GPUs to allow faster model training and prediction. DeepForest is a PyTorch Lightning module, which automatically distributes data to available GPUs. If using a release model with training, the module can be moved from CPU to GPU for prediction using the pytorch.to() method.

from deepforest import main
m = main.deepforest()
m.load_model("weecology/deepforest-tree")
print("Current device is {}".format(m.device))
m.to("cuda")
print("Current device is {}".format(m.device))

Current device is cpu
Current device is cuda:0

Distributed multi-GPU prediction outside of the training module is not yet implemented. We welcome pull requests for additional support.