lightweight_mmm
lightweight_mmm copied to clipboard
google
When running the imports, I get the following error message related to numba/numpy dependency: ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. numba 0.56.4 requires numpy<1.24,>=1.18, but you have numpy 1.24.2 which is incompatible.
When trying to downgrade numpy's version, I get another error stating that lightweight-mmm-0.1.7 needs the latest numpy version. Appreciate the help
What worked for me, after a morning of debugging - uninstall numpy, setuptools and sklearn. Install scikit-learn, force setuptools to version 64.0.2 (required by some libraries lightweight_mmm relies on) and force lightweight_mmm to version 0.1.5.
Tested with Google Colab, on a fresh runtime.
You can't go below 0.1.5 with lmmm because prior versions used sklearn instead of scikit-learn, and that package has been permanently yanked from pypi. You may get around this with aliasing or some other dark magics, but I don't see the point.
%pip uninstall -y numpy
%pip uninstall -y setuptools
%pip uninstall -y sklearn
%pip install -U scikit-learn
%pip install setuptools==64.0.2
%pip install lightweight_mmm==0.1.5
pip install lightweight_mmm==0.1.5
ERROR: Could not find a version that satisfies the requirement jaxlib>=0.3.14 (from lightweight-mmm) (from versions: none)
ERROR: No matching distribution found for jaxlib>=0.3.14
