AgentTorch is meant to be used in a Python 3.9 environment. If you have not installed Python 3.9, please do so first from python.org/downloads.
To install the project, run:
> pip install git+https://github.com/agenttorch/agenttorch
To run some models, you may need to separately install their dependencies. These
usually include torch
,
torch_geometric
,
and osmnx
.
For the sake of completeness, a summary of the commands required is given below:
# on macos, cuda is not available:
> pip install torch torchvision torchaudio
> pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv
> pip install osmnx
# on ubuntu, where ${CUDA} is the cuda version:
> pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/${CUDA}
> pip install pyg_lib torch_scatter torch_sparse torch_cluster torch_spline_conv -f https://data.pyg.org/whl/torch-2.0.0+${CUDA}.html
> pip install osmnx
The code has been tested on macOS Catalina 10.1.7 and Ubuntu 22.04.2 LTS. Large-scale experiments are run using Nvidia's TitanX and V100 GPUs.