Environment setup

The simplest way to get an appropriate environment set up for these tutorials is using a conda environment.

If you are an ISIS facility users, a fast “mamba”-enable conda impelementation is available on IDAaaS. Otherwise, if you are not sure what is available it may be best to use Miniforge or micromamba to get up and running.

An environment file is provided in this repository; create an environment and activate it with e.g.

mamba env create -f env.yaml
mamba activate abins-tutorials

Be aware that this environment has about 3GB of depenencies; this is dominated by the NVIDIA/CUDA libraries needed to run on GPU. If you are not using GPU, consider removing the “pytorch-cuda” line before installing!