Hello Amirashkan-
I am assuming you are using the energysystem branch of the code. An alternative to using Conda and the config.yml file that conda uses is to use pip to download the dependencies from PyPI. I’m not sure why the .yml isn’t working for you. The versions specified are quite dated and probably not supported (via conda) for M1.
I have the project energysystem branch running on M2 architecture using the recipe below. As an aside, I’m midway through many upgrades that will bring the whole project up to current python/pyomo versions, but until then this should work:
1. Install your own version of python from
python.org by looking through the earlier available builds. The build I used is noted in the comments in the attached file
2. Create a virtual environment that uses this specific version of python, call it whatever, I call it “venv37” so I know the core python version
$: python3.7 -m venv venv37
3. Activate that venv (there are many tutorials on using the venv module online if this is confusing)
$: source venv37/bin/activate
4. While inside the venv, use pip to install everything in the attached requirements.txt file (the command is shown in the file)
5. As noted in the file, you will also need a solver on your machine and that should be installed separately as well if you don’t have one. I recommend ‘cbc’ as a starter. There are install instructions on this page:
https://github.com/coin-or/Cbc. I see that there is a conda and hombebrew option for Mac. I’m a homebrew user and that works.
Most common IDE’s have tools to automate much of this, but you’ll still need to go to
python.org to get the base version of python installed, and the solver.
Best,
Jeff