Some users have had an issue with Conda failing whilst creating the environment, giving the following message:

Collecting package metadata (repodata.json): - Killed

This is due to Conda running out of RAM during install, documented here. This is in line with problems found a few months ago, with 8 GB of RAM often not being sufficient. Unfortunately Conda isn't very efficient with memory, and there is little we can do about this.

I have updated the docs to include 3 options to fix this issue:

• Increase available RAM (if running on a compute cluster etc)
• Use Mamba, a more efficient drop-in replacement for Conda. This is what I personally use locally and find it to be vastly better — however, I can appreciate this isn't an option in some situations!
• Create a blank conda environment and install the required packages:

  conda create --name sharpy python=3.10
  conda config –add channels conda-forge
  conda install eigen libopenblas libblas libcblas liblapack libgfortran libgcc libgfortran-ng

Thank you to the students who both found this issue and documented the third fix method!